Search results for: machine translation (MT)

Authors: Lavanya Annadi, Fahimeh Jafari

Abstract:

Transportation brings immense benefits to society, but it also has its costs. Costs include such as the cost of infrastructure, personnel and equipment, but also the loss of life and property in traffic accidents on the road, delays in travel due to traffic congestion and various indirect costs in terms of air transport. More research has been done to identify the various factors that affect road accidents, such as road infrastructure, traffic, sociodemographic characteristics, land use, and the environment. The aim of this research is to predict the probabilistic crash prediction of vehicles using machine learning due to natural and structural reasons by excluding spontaneous reasons like overspeeding etc., in the United States. These factors range from weather factors, like weather conditions, precipitation, visibility, wind speed, wind direction, temperature, pressure, and humidity to human made structures like road structure factors like bump, roundabout, no exit, turning loop, give away, etc. Probabilities are dissected into ten different classes. All the predictions are based on multiclass classification techniques, which are supervised learning. This study considers all crashes that happened in all states collected by the US government. To calculate the probability, multinomial expected value was used and assigned a classification label as the crash probability. We applied three different classification models, including multiclass Logistic Regression, Random Forest and XGBoost. The numerical results show that XGBoost achieved a 75.2% accuracy rate which indicates the part that is being played by natural and structural reasons for the crash. The paper has provided in-deep insights through exploratory data analysis.

Keywords: road safety, crash prediction, exploratory analysis, machine learning

Procedia PDF Downloads 108

Authors: Luis C. Parra

Abstract:

The significant wave height prediction is an issue of great interest in the field of coastal activities because of the non-linear behavior of the wave height and its complexity of prediction. This study aims to present a machine learning model to forecast the significant wave height of the oceanographic wave measuring buoys anchored at Mooloolaba of the Queensland Government Data. Modeling was performed by a multilayer perceptron neural network-genetic algorithm (GA-MLP), considering Relu(x) as the activation function of the MLPNN. The GA is in charge of optimized the MLPNN hyperparameters (learning rate, hidden layers, neurons, and activation functions) and wrapper feature selection for the window width size. Results are assessed using Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). The GAMLPNN algorithm was performed with a population size of thirty individuals for eight generations for the prediction optimization of 5 steps forward, obtaining a performance evaluation of 0.00104 MSE, 0.03222 RMSE, 0.02338 MAE, and 0.71163% of MAPE. The results of the analysis suggest that the MLPNNGA model is effective in predicting significant wave height in a one-step forecast with distant time windows, presenting 0.00014 MSE, 0.01180 RMSE, 0.00912 MAE, and 0.52500% of MAPE with 0.99940 of correlation factor. The GA-MLP algorithm was compared with the ARIMA forecasting model, presenting better performance criteria in all performance criteria, validating the potential of this algorithm.

Keywords: significant wave height, machine learning optimization, multilayer perceptron neural networks, evolutionary algorithms

Procedia PDF Downloads 104

Authors: Merve Yünlü, Nihat Kandemir, Aylin Ersoy

Abstract:

Temperature sensors are used in home appliances not only to monitor the basic functions of the machine but also to minimize energy consumption and ensure safe operation. In parallel with the development of smart home applications and IoT algorithms, these sensors produce important data such as the frequency of use of the machine, user preferences, and the compilation of critical data in terms of diagnostic processes for fault detection throughout an appliance's operational lifespan. Commercially available thin-film resistive temperature sensors have a well-established manufacturing procedure that allows them to operate over a wide temperature range. However, these sensors are over-designed for white goods applications. The operating temperature range of these sensors is between -70°C and 850°C, while the temperature range requirement in home appliance applications is between 23°C and 500°C. To ensure the operation of commercial sensors in this wide temperature range, usually, a platinum coating of approximately 1-micron thickness is applied to the wafer. However, the use of platinum in coating and the high coating thickness extends the sensor production process time and therefore increases sensor costs. In this study, an attempt was made to develop a low-cost temperature sensor design and production method that meets the technical requirements of white goods applications. For this purpose, a custom design was made, and design parameters (length, width, trim points, and thin film deposition thickness) were optimized by using statistical methods to achieve the desired resistivity value. To develop thin film resistive temperature sensors, one side polished sapphire wafer was used. To enhance adhesion and insulation 100 nm silicon dioxide was coated by inductively coupled plasma chemical vapor deposition technique. The lithography process was performed by a direct laser writer. The lift-off process was performed after the e-beam evaporation of 10 nm titanium and 280 nm platinum layers. Standard four-point probe sheet resistance measurements were done at room temperature. The annealing process was performed. Resistivity measurements were done with a probe station before and after annealing at 600°C by using a rapid thermal processing machine. Temperature dependence between 25-300 °C was also tested. As a result of this study, a temperature sensor has been developed that has a lower coating thickness than commercial sensors but can produce reliable data in the white goods application temperature range. A relatively simplified but optimized production method has also been developed to produce this sensor.

Keywords: thin film resistive sensor, temperature sensor, household appliance, sustainability, energy efficiency

Procedia PDF Downloads 69

Authors: S. Chee Choy, Pauline Swee Choo Goh, Yow Lin Liew

Abstract:

The purpose of this study is to evaluate the English version and a Malay translation of the 21-item Learner Awareness Questionnaire for its application to assess student learning in higher education. The Learner Awareness Questionnaire, originally written in English, is a quantitative measure of how and why students learn. The questionnaire gives an indication of the process and motives to learn using four scales: survival, establishing stability, approval, and loving to learn. Data in the present study came from 680 university students enrolled in various programs in Malaysia. The Malay version of the questionnaire supported a similar four-factor structure and internal consistency to the English version. The four factors of the Malay version also showed moderate to strong correlations with those of the English versions. The results suggest that the Malay version of the questionnaire is similar to the English version. However, further refinement for the questions is needed to strengthen the correlations between the two questionnaires.

Keywords: student learning, learner awareness, questionnaire development, instrument validation

Procedia PDF Downloads 423

Authors: Mark Ashton Newman, Richard Blagrove, Jonathan Folland

Abstract:

The force-velocity profile of an individual has been shown to influence success in ballistic movements, independent of the individuals' maximal power output; therefore, effective and accurate evaluation of an individual’s F-V characteristics and not solely maximal power output is important. The relatively narrow range of loads typically utilised during force-velocity profiling protocols due to the difficulty in obtaining force data at high velocities may bring into question the accuracy of the F-V slope along with predictions pertaining to the maximum force that the system can produce at a velocity of null (F₀) and the theoretical maximum velocity against no load (V₀). As such, the reliability of the slope of the force-velocity profile, as well as V₀, has been shown to be relatively poor in comparison to F₀ and maximal power, and it has been recommended to assess velocity at loads closer to both F₀ and V₀. The aim of the present study was to assess the relative and absolute reliability of an instrumented novel leg press machine which enables the assessment of force and velocity data at loads equivalent to ≤ 10% of one repetition maximum (1RM) through to 1RM during a ballistic leg press movement. The reliability of maximal and mean force, velocity, and power, as well as the respective force-velocity and power-velocity relationships and the linearity of the force-velocity relationship, were evaluated. Sixteen male strength-trained individuals (23.6 ± 4.1 years; 177.1 ± 7.0 cm; 80.0 ± 10.8 kg) attended four sessions; during the initial visit, participants were familiarised with the leg press, modified to include a mounted force plate (Type SP3949, Force Logic, Berkshire, UK) and a Micro-Epsilon WDS-2500-P96 linear positional transducer (LPT) (Micro-Epsilon, Merseyside, UK). Peak isometric force (IsoMax) and a dynamic 1RM, both from a starting position of 81% leg length, were recorded for the dominant leg. Visits two to four saw the participants carry out the leg press movement at loads equivalent to ≤ 10%, 30%, 50%, 70%, and 90% 1RM. IsoMax was recorded during each testing visit prior to dynamic F-V profiling repetitions. The novel leg press machine used in the present study appears to be a reliable tool for measuring F and V-related variables across a range of loads, including velocities closer to V₀ when compared to some of the findings within the published literature. Both linear and polynomial models demonstrated good to excellent levels of reliability for SFV and F₀ respectively, with reliability for V₀ being good using a linear model but poor using a 2nd order polynomial model. As such, a polynomial regression model may be most appropriate when using a similar unilateral leg press setup to predict maximal force production capabilities due to only a 5% difference between F₀ and obtained IsoMax values with a linear model being best suited to predict V₀.

Keywords: force-velocity, leg-press, power-velocity, profiling, reliability

Procedia PDF Downloads 55

Authors: Shraddha Jain Sharma, Ratnalata Gupta

Abstract:

Predicting stock movements in the financial market is both difficult and rewarding. Analysts and academics are increasingly using advanced approaches such as machine learning techniques to anticipate stock price patterns, thanks to the expanding capacity of computing and the recent advent of graphics processing units and tensor processing units. Stock market prediction is a type of time series prediction that is incredibly difficult to do since stock prices are influenced by a variety of financial, socioeconomic, and political factors. Furthermore, even minor mistakes in stock market price forecasts can result in significant losses for companies that employ the findings of stock market price prediction for financial analysis and investment. Soft computing techniques are increasingly being employed for stock market prediction due to their better accuracy than traditional statistical methodologies. The proposed research looks at the need for soft computing techniques in stock market prediction, the numerous soft computing approaches that are important to the field, past work in the area with their prominent features, and the significant problems or issue domain that the area involves. For constructing a predictive model, the major focus is on neural networks and fuzzy logic. The stock market is extremely unpredictable, and it is unquestionably tough to correctly predict based on certain characteristics. This study provides a complete overview of the numerous strategies investigated for high accuracy prediction, with a focus on the most important characteristics.

Keywords: stock market prediction, artificial intelligence, artificial neural networks, fuzzy logic, accuracy, deep learning, machine learning, stock price, trading volume

Procedia PDF Downloads 87

Authors: Justine Cyril E. Nunag, Nestor B. Sabado Jr., Jienne Chester M. Tolosa

Abstract:

Concrete has high compressive strength but a low-tension strength. The small tensile strength of concrete is regarded as its primary weakness, which is why it is typically reinforced with steel, a material that is resistant to tension. Even with steel, however, cracking can occur. In strengthening concrete, only a few researchers have modified the water to be used in a concrete mix. This study aims to compare the split tensile strength of normal structural concrete to concrete prepared with magnetic water and a quick setting admixture. In this context, magnetic water is defined as tap water that has undergone a magnetic process to become magnetized water. To test the hypothesis that magnetized concrete leads to higher split tensile strength, twenty concrete specimens were made. There were five groups, each with five samples, that were differentiated by the number of cycles (0, 50, 100, and 150). The data from the Universal Testing Machine's split tensile strength were then analyzed using various statistical models and tests to determine the significant effect of magnetized water. The result showed a moderate (+0.579) but still significant degree of correlation. The researchers also discovered that using magnetic water for 50 cycles did not result in a significant increase in the concrete's split tensile strength, which influenced the analysis of variance. These results suggest that a concrete mix containing magnetic water and a quick-setting admixture alters the typical split tensile strength of normal concrete. Magnetic water has a significant impact on concrete tensile strength. The hardness property of magnetic water influenced the split tensile strength of concrete. In addition, a higher number of cycles results in a strong water magnetism. The laboratory test results show that a higher cycle translates to a higher tensile strength.

Keywords: hardness property, magnetic water, quick-setting admixture, split tensile strength, universal testing machine

Procedia PDF Downloads 143

Authors: Muogbo Chukwudi Peter, Gbabo Agidi

Abstract:

Influence of different turmeric rhizome sizes and machine operational speed on the field capacity and efficiency of a developed prototype tractor-drawn turmeric planter was studied. This was done with a view to ascertaining how the field capacity and field efficiency were affected by the turmeric rhizome lengths and tractor operational speed. The turmeric rhizome planter consists of trapezoidal hopper, grooved cylindrical metering devise, rectangular frame, ground wheels made of mild steel, furrow opener, chain/sprocket drive system, three linkage point seed delivery tube and press wheel. The experiment was randomized in a factorial design of three levels of rhizome lengths (30, 45 and 60 mm) and operational speeds of 8, 10, and 12 kmh-1. About 3 kg cleaned turmeric rhizomes were introduced into each hopper of the planter and were planted 30 m2 of experimental plot. During the field evaluation of the planter, the effective field capacity, field efficiency, missing index, multiple index and percentage rhizome bruise were evaluated. 30.08% was recorded for maximum percentage bruise on the rhizome. The mean effective field capacity ranged between 0.63 – 0.96hah-1 at operational speeds of 8 and 12kmh-1 respectively and 45 mm rhizome length. The result also shows that the mean efficiency was obtained to be 65.8%. The percentage rhizome bruise decreases with increase in operational speed. The highest and lowest percentage turmeric rhizome miss index of 35% were recorded for turmeric rhizome length of 30 mm at a speed of 10 kmhr-1 and 8 kmhr-1, respectively. The potential implications of the experimental result is to determine the optimal machine process conditions for higher field capacity and gross reduction in mechanical injury (bruise) of planted turmeric rhizomes.

Keywords: rhizome sizes, operational speed, field capacity. field efficiency, turmeric rhizome, planter

Procedia PDF Downloads 59

Authors: Priyanka Jain, Ashish K. Sharma, Esha Shukla, K. J. Mukherjee

Abstract:

We propose a paradigm shift in our approach to design improved platforms for recombinant protein production, by addressing system level issues rather than the individual steps associated with recombinant protein synthesis like transcription, translation, etc. We demonstrate that by controlling and modulating the cellular stress response (CSR), which is responsible for feedback control of protein synthesis, we can generate hyper-producing strains. We did transcriptomic profiling of post-induction cultures, expressing different types of protein, to analyze the nature of this cellular stress response. We found significant down-regulation of substrate utilization, translation, and energy metabolism genes due to generation CSR inside the host cell. However, transcription profiling has also shown that many genes are up-regulated post induction and their role in modulating the CSR is unclear. We hypothesized that these up-regulated genes trigger signaling pathways, generating the CSR and concomitantly reduce the recombinant protein yield. To test this hypothesis, we knocked out the up-regulated genes, which did not have any downstream regulatees, and analyzed their impact on cellular health and recombinant protein expression. Two model proteins i.e., GFP and L-Asparaginase were chosen for this analysis. We observed a significant improvement in expression levels, with some knock-outs showing more than 7-fold higher expression compared to control. The 10 best single knock-outs were chosen to make 45 combinations of all possible double knock-outs. A further increase in expression was observed in some of these double knock- outs with GFP levels being highest in a double knock-out ΔyhbC + ΔelaA. However, for L-Asparaginase which is a secretory protein, the best results were obtained using a combination of ΔelaA+ΔcysW knock-outs. We then tested all the knock outs for their ability to enhance the expression of a 'difficult-to-express' protein. The Rubella virus E1 protein was chosen and tagged with sfGFP at the C-terminal using a linker peptide for easy online monitoring of expression of this fusion protein. Interestingly, the highest increase in Rubella-sGFP levels was obtained in the same double knock-out ΔelaA + ΔcysW (5.6 fold increase in expression yield compared to the control) which gave the highest expression for L-Asparaginase. However, for sfGFP alone, the ΔyhbC+ΔmarR knock-out gave the highest level of expression. These results indicate that there is a fair degree of commonality in the nature of the CSR generated by the induction of different proteins. Transcriptomic profiling of the double knock out showed that many genes associated with the translational machinery and energy biosynthesis did not get down-regulated post induction, unlike the control where these genes were significantly down-regulated. This confirmed our hypothesis of these genes playing an important role in the generation of the CSR and allowed us to design a strategy for making better expression hosts by simply knocking out key genes. This strategy is radically superior to the previous approach of individually up-regulating critical genes since it blocks the mounting of the CSR thus preventing the down-regulation of a very large number of genes responsible for sustaining the flux through the recombinant protein production pathway.

Keywords: cellular stress response, GFP, knock-outs, up-regulated genes

Procedia PDF Downloads 224

Authors: Ravneet Kaur, Joydeep Acharya, Sudhanshu Gaur

Abstract:

Artificial Intelligence (AI) has the potential to transform manufacturing by improving shop floor processes such as production, maintenance and quality. However, industrial datasets are notoriously difficult to extract in a real-time, streaming fashion thus, negating potential AI benefits. The main example is some specialized industrial controllers that are operated by custom software which complicates the process of connecting them to an Information Technology (IT) based data acquisition network. Security concerns may also limit direct physical access to these controllers for data acquisition. To connect the Operational Technology (OT) data stored in these controllers to an AI application in a secure, reliable and available way, we propose a novel Industrial IoT (IIoT) solution in this paper. In this solution, we demonstrate how video cameras can be installed in a factory shop floor to continuously obtain images of the controller HMIs. We propose image pre-processing to segment the HMI into regions of streaming data and regions of fixed meta-data. We then evaluate the performance of multiple Optical Character Recognition (OCR) technologies such as Tesseract and Google vision to recognize the streaming data and test it for typical factory HMIs and realistic lighting conditions. Finally, we use the meta-data to match the OCR output with the temporal, domain-dependent context of the data to improve the accuracy of the output. Our IIoT solution enables reliable and efficient data extraction which will improve the performance of subsequent AI applications.

Keywords: human machine interface, industrial internet of things, internet of things, optical character recognition, video analytics

Procedia PDF Downloads 107

Authors: Adriana Haulica

Abstract:

Powered by Machine Learning, Precise medicine is tailored by now to use genetic and molecular profiling, with the aim of optimizing the therapeutic benefits for cohorts of patients. As the majority of Machine Language algorithms come from heuristics, the outputs have contextual validity. This is not very restrictive in the sense that medicine itself is not an exact science. Meanwhile, the progress made in Molecular Biology, Bioinformatics, Computational Biology, and Precise Medicine, correlated with the huge amount of human biology data and the increase in computational power, opens new healthcare challenges. A more accurate diagnosis is needed along with real-time treatments by processing as much as possible from the available information. The purpose of this paper is to present a deeper vision for the future of Artificial Intelligence in Precise medicine. In fact, actual Machine Learning algorithms use standard mathematical knowledge, mostly Euclidian metrics and standard computation rules. The loss of information arising from the classical methods prevents obtaining 100% evidence on the diagnosis process. To overcome these problems, we introduce MEDICOMPILLS, a new architectural concept tool of information processing in Precise medicine that delivers diagnosis and therapy advice. This tool processes poly-field digital resources: global knowledge related to biomedicine in a direct or indirect manner but also technical databases, Natural Language Processing algorithms, and strong class optimization functions. As the name suggests, the heart of this tool is a compiler. The approach is completely new, tailored for omics and clinical data. Firstly, the intrinsic biological intuition is different from the well-known “a needle in a haystack” approach usually used when Machine Learning algorithms have to process differential genomic or molecular data to find biomarkers. Also, even if the input is seized from various types of data, the working engine inside the MEDICOMPILLS does not search for patterns as an integrative tool. This approach deciphers the biological meaning of input data up to the metabolic and physiologic mechanisms, based on a compiler with grammars issued from bio-algebra-inspired mathematics. It translates input data into bio-semantic units with the help of contextual information iteratively until Bio-Logical operations can be performed on the base of the “common denominator “rule. The rigorousness of MEDICOMPILLS comes from the structure of the contextual information on functions, built to be analogous to mathematical “proofs”. The major impact of this architecture is expressed by the high accuracy of the diagnosis. Detected as a multiple conditions diagnostic, constituted by some main diseases along with unhealthy biological states, this format is highly suitable for therapy proposal and disease prevention. The use of MEDICOMPILLS architecture is highly beneficial for the healthcare industry. The expectation is to generate a strategic trend in Precise medicine, making medicine more like an exact science and reducing the considerable risk of errors in diagnostics and therapies. The tool can be used by pharmaceutical laboratories for the discovery of new cures. It will also contribute to better design of clinical trials and speed them up.

Keywords: bio-semantic units, multiple conditions diagnosis, NLP, omics

Procedia PDF Downloads 63

Authors: Adam J. Bowen

Abstract:

Abstract— Should robots have rights or legal protections. Often debates concerning whether robots and AI should be afforded rights focus on conditions of personhood and the possibility of future advanced forms of AI satisfying particular intrinsic cognitive and moral attributes of rights-holding persons. Such discussions raise compelling questions about machine consciousness, autonomy, and value alignment with human interests. Although these are important theoretical concerns, especially from a future design perspective, they provide limited guidance for addressing the moral and legal standing of current and near-term AI that operate well below the cognitive and moral agency of human persons. Robots and AI are already being pressed into service in a wide range of roles, especially in healthcare and biomedical contexts. The design and large-scale implementation of robots in the context of core societal institutions like healthcare systems continues to rapidly develop. For example, we bring them into our homes, hospitals, and other care facilities to assist in care for the sick, disabled, elderly, children, or otherwise vulnerable persons. We enlist surgical robotic systems in precision tasks, albeit still human-in-the-loop technology controlled by surgeons. We also entrust them with social roles involving companionship and even assisting in intimate caregiving tasks (e.g., bathing, feeding, turning, medicine administration, monitoring, transporting). There have been advances to enable severely disabled persons to use robots to feed themselves or pilot robot avatars to work in service industries. As the applications for near-term AI increase and the roles of robots in restructuring our biomedical practices expand, we face pressing questions about the normative implications of human-robot interactions and collaborations in our collective worldmaking, as well as the moral and legal status of robots. This paper argues that robots operating in public and private spaces be afforded some protections as either moral patients or legal agents to establish prohibitions on robot abuse, misuse, and mistreatment. We already implement robots and embed them in our practices and institutions, which generates a host of human-to-machine and machine-to-machine relationships. As we interact with machines, whether in service contexts, medical assistance, or home health companions, these robots are first encountered in relationship to us and our respective roles in the encounter (e.g., surgeon, physical or occupational therapist, recipient of care, patient’s family, healthcare professional, stakeholder). This proposal aims to outline a framework for establishing limiting factors and determining the extent of moral or legal protections for robots. In doing so, it advocates for a relational approach that emphasizes the priority of mapping the complex contextually sensitive roles played and the relations in which humans and robots stand to guide policy determinations by relevant institutions and authorities. The relational approach must also be technically informed by the intended uses of the biomedical technologies in question, Design History Files, extensive risk assessments and hazard analyses, as well as use case social impact assessments.

Keywords: biomedical robots, robot ethics, robot laws, human-robot interaction

Procedia PDF Downloads 114

Authors: Yuanchao Sang, Zhihao Gong, Longsheng Chen, Long Chen

Abstract:

Clothing fashion represents a human’s aesthetic appreciation towards everyday outfits and appetite for fashion, and it reflects the development of status in society, humanity, and economics. However, modelling fashion by machine is extremely challenging because fashion is too abstract to be efficiently described by machines. Even human beings can hardly reach a consensus about fashion. In this paper, we are dedicated to answering a fundamental fashion-related problem: what image feature best describes clothing fashion? To address this issue, we have designed and evaluated various image features, ranging from traditional low-level hand-crafted features to mid-level style awareness features to various current popular deep neural network-based features, which have shown state-of-the-art performance in various vision tasks. In summary, we tested the following 9 feature representations: color, texture, shape, style, convolutional neural networks (CNNs), CNNs with distance metric learning (CNNs&DML), AutoEncoder, CNNs with multiple layer combination (CNNs&MLC) and CNNs with dynamic feature clustering (CNNs&DFC). Finally, we validated the performance of these features on two publicly available datasets. Quantitative and qualitative experimental results on both intra-domain and inter-domain fashion clothing image retrieval showed that deep learning based feature representations far outweigh traditional hand-crafted feature representation. Additionally, among all deep learning based methods, CNNs with explicit feature clustering performs best, which shows feature clustering is essential for discriminative fashion feature representation.

Keywords: convolutional neural network, feature representation, image processing, machine modelling

Procedia PDF Downloads 136

Authors: Jacob Livingston Slosser, Naja Holten Moller, Thomas Troels Hildebrandt, Henrik Palmer Olsen

Abstract:

In this paper, we propose a model for human-AI interaction in public administration that involves legal decision-making. Inspired by Alan Turing’s test for machine intelligence, we propose a way of institutionalizing a continuous working relationship between man and machine that aims at ensuring both good legal quality and higher efficiency in decision-making processes in public administration. We also suggest that our model enhances the legitimacy of using AI in public legal decision-making. We suggest that case loads in public administration could be divided between a manual and an automated decision track. The automated decision track will be an algorithmic recommender system trained on former cases. To avoid unwanted feedback loops and biases, part of the case load will be dealt with by both a human case worker and the automated recommender system. In those cases an experienced human case worker will have the role of an evaluator, choosing between the two decisions. This model will ensure that the algorithmic recommender system is not compromising the quality of the legal decision making in the institution. It also enhances the legitimacy of using algorithmic decision support because it provides justification for its use by being seen as superior to human decisions when the algorithmic recommendations are preferred by experienced case workers. The paper outlines in some detail the process through which such a model could be implemented. It also addresses the important issue that legal decision making is subject to legislative and judicial changes and that legal interpretation is context sensitive. Both of these issues requires continuous supervision and adjustments to algorithmic recommender systems when used for legal decision making purposes.

Keywords: administrative law, algorithmic decision-making, decision support, public law

Procedia PDF Downloads 214

Authors: Kai-Jui Kou, Tzu-Ling Shen, Ying-Fang Wang

Abstract:

The objectives of the present study are to characterize nanoparticles generated from the different working type in 3D printing room and the stack flue during 3D printing process. The studied laboratory (10.5 m× 7.2 m × 3.2 m) with a ventilation rate of 500 m³/H is installed a 3D metal printing machine. Direct-reading instrument of a scanning mobility particle sizer (SMPS, Model 3082, TSI Inc., St. Paul, MN, USA) was used to conduct static sampling for nanoparticle number concentration and particle size distribution measurements. The SMPS obtained particle number concentration at every 3 minutes, the diameter of the SMPS ranged from 11~372 nm when the aerosol and sheath flow rates were set at 0.6 and 6 L/min, respectively. The concentrations of background, printing process, clearing operation, and screening operation were performed in the laboratory. On the other hand, we also conducted nanoparticle measurement on the 3D printing machine's stack flue to understand its emission characteristics. Results show that the nanoparticles emitted from the different operation process were the same distribution in the form of the uni-modal with number median diameter (NMD) as approximately 28.3 nm to 29.6 nm. The number concentrations of nanoparticles were 2.55×10³ count/cm³ in laboratory background, 2.19×10³ count/cm³ during printing process, 2.29×10³ count/cm³ during clearing process, 3.05×10³ count/cm³ during screening process, 2.69×10³ count/cm³ in laboratory background after printing process, and 6.75×10³ outside laboratory, respectively. We found that there are no emission nanoparticles during the printing process. However, the number concentration of stack flue nanoparticles in the ongoing print is 1.13×10⁶ count/cm³, and that of the non-printing is 1.63×10⁴ count/cm³, with a NMD of 458 nm and 29.4 nm, respectively. It can be confirmed that the measured particle size belongs to easily penetrate the filter in theory during the printing process, even though the 3D printer has a high-efficiency filtration device. Therefore, it is recommended that the stack flue of the 3D printer would be equipped with an appropriate dust collection device to prevent the operators from exposing these hazardous particles.

Keywords: nanoparticle, particle emission, 3D printing, number concentration

Procedia PDF Downloads 179

Authors: Wen-Chung Chiang, Lun-Jou Lo, Hsiu-Hsia Lin

Abstract:

This study aimed to improve the evaluation of facial symmetry, which is crucial for planning and assessing outcomes in orthognathic surgery (OGS). Facial symmetry plays a key role in both aesthetic and functional aspects of OGS, making its accurate evaluation essential for optimal surgical results. To address the limitations of traditional methods, a different approach was developed, combining three-dimensional (3D) facial contour mapping with hyperdimensional (HD) computing to enhance precision and efficiency in symmetry assessments. The study was conducted at Chang Gung Memorial Hospital, where data were collected from 2018 to 2023 using 3D cone beam computed tomography (CBCT), a highly detailed imaging technique. A large and comprehensive dataset was compiled, consisting of 150 normal individuals and 2,800 patients, totaling 5,750 preoperative and postoperative facial images. These data were critical for training a machine learning model designed to analyze and quantify facial symmetry. The machine learning model was trained to process 3D contour data from the CBCT images, with HD computing employed to power the facial symmetry quantification system. This combination of technologies allowed for an objective and detailed analysis of facial features, surpassing the accuracy and reliability of traditional symmetry assessments, which often rely on subjective visual evaluations by clinicians. In addition to developing the system, the researchers conducted a retrospective review of 3D CBCT data from 300 patients who had undergone OGS. The patients’ facial images were analyzed both before and after surgery to assess the clinical utility of the proposed system. The results showed that the facial symmetry algorithm achieved an overall accuracy of 82.5%, indicating its robustness in real-world clinical applications. Postoperative analysis revealed a significant improvement in facial symmetry, with an average score increase of 51%. The mean symmetry score rose from 2.53 preoperatively to 3.89 postoperatively, demonstrating the system's effectiveness in quantifying improvements after OGS. These results underscore the system's potential for providing valuable feedback to surgeons and aiding in the refinement of surgical techniques. The study also led to the development of a web-based system that automates facial symmetry assessment. This system integrates HD computing and 3D contour mapping into a user-friendly platform that allows for rapid and accurate evaluations. Clinicians can easily access this system to perform detailed symmetry assessments, making it a practical tool for clinical settings. Additionally, the system facilitates better communication between clinicians and patients by providing objective, easy-to-understand symmetry scores, which can help patients visualize the expected outcomes of their surgery. In conclusion, this study introduced a valuable and highly effective approach to facial symmetry evaluation in OGS, combining 3D contour mapping, HD computing, and machine learning. The resulting system achieved high accuracy and offers a streamlined, automated solution for clinical use. The development of the web-based platform further enhances its practicality, making it a valuable tool for improving surgical outcomes and patient satisfaction in orthognathic surgery.

Keywords: facial symmetry, orthognathic surgery, facial contour mapping, hyperdimensional computing

Procedia PDF Downloads 3

Authors: John Q. Todd

Abstract:

Given that modern equipment can provide comprehensive health, status, and error condition data via built-in sensors, maintenance organizations have a new and valuable source of insight to take advantage of. This presentation will expose what these data payloads might look like and how they can be filtered, visualized, calculated into metrics, used for machine learning, and generate alerts for further action.

Keywords: condition based maintenance, equipment data, metrics, alerts

Procedia PDF Downloads 184

Authors: Muhammad Saleem

Abstract:

The research paper entitled above would be a comprehensive and critical study of translations of the meanings of the Holy Qur’an. The discussion will deal with the targets & purposes of Urdu (National Language of Pakistan) translators of the meanings of the Holy Qur’an. There are more than 400 translations of the meanings of the Holy Qur’an in the Urdu Language. Muslims, non-Muslims and some organizations have made translations of the meanings of the Holy Qur’an to meet various targets. It is observed that all Urdu translators have not translated the Qur’an with a single objective and motivation; rather, some are biased and strive to discredit the Qur’an. Thus, they have made unauthentic and fabricated translations of the Qur’an. Some optimistically believe that they intend to do a service, whereas others pessimistically hold that they treacherously seek to further their rule. Some of them have been observed to be against Islam, starting their activities with spite, but after perceiving the truths of Islam and the miracle and greatness of the Holy Qur’an, they submitted to Islam, embracing it with pure hearts. Some translators made their translations of the meanings of the Holy Qur’an to serve Allah, and some of them have done their translations to earn only. All these translations vary from one to another due to style, trend, type, method and style. Some Urdu translations have been made to fulfill the lingual requirements. Some translations have been made by Muslim scholars to reduce the influence of Urdu translations of the meanings of the Holy Qur’an by Non-Muslims. The article deals with the various purposes of the translators of the meanings of the Holy Qur’an.

Keywords: Qur'an, translation, urdu, language

Procedia PDF Downloads 34

Authors: Andrew Laghos

Abstract:

Computer-assisted language learning (CALL) is defined as the use of computers to help learn languages. In this study we look at several different types of CALL tools and applications and how they can assist Adults and Young Learners in learning the English language as a foreign, second or additional language. It is important to identify the roles of the teacher and the learners, and what the learners’ motivations are for learning the language. Audio, video, interactive multimedia games, online translation services, conferencing, chat rooms, discussion forums, social networks, social media, email communication, songs and music video clips are just some of the many ways computers are currently being used to enhance language learning. CALL may be used for classroom teaching as well as for online and mobile learning. Advantages and disadvantages of CALL are discussed and the study ends with future predictions of CALL.

Keywords: computer-assisted language learning (CALL), teaching English as a foreign language (TEFL/EFL), adult learners, young learners

Procedia PDF Downloads 431

Authors: Manssour Habbash, Srinivasa Rao Idapalapati

Abstract:

Although writing is a multifaceted task, teaching writing is a demanding task basically for two reasons: Grammar and Syntax. This article provides a method of teaching writing that was found to be effective in improving students’ academic writing composition skill. The article explains the concepts of ‘guided-discovery’ and ‘guided-construction’ upon which a method of teaching writing is grounded and developed. Providing a brief commentary on what the core could mean primarily, the article presents an exposition of understanding and identifying the core and building upon the core that can demonstrate the way a teacher can make use of the concepts in teaching for improving the writing skills of their students. The method is an adaptation of grammar translation method that has been improvised to suit to a student-centered classroom environment. An intervention of teaching writing through this method was tried out with positive outcomes in formal classroom research setup, and in view of the content’s quality that relates more to the classroom practices and also in consideration of its usefulness to the practicing teachers the process and the findings are presented in a narrative form along with the results in tabular form.

Keywords: core of a text, guided construction, guided discovery, theme of a text

Procedia PDF Downloads 376

Authors: Piyapat Chitpirom, Teerakiat Jareonsettasin, Chintida Vichitsophaphan

Abstract:

The Foundation of Virtuous Youth was established and supported by the Crown Property Bureau, with the intention to instill goodness in Thai youth. The Centre for Educational Psychology is one of the three units under the foundation. We aim to develop programmes that can be used to improve the quality of education in schools. Translation of the King’s message in keeping with the modern research from various sources, our team create 6 programmes: (1) Teacher-Student Relationship (2) Growth Mindset (3) Socratic Teaching (4) Peer Tutoring (5) Parental Involvement (6) Inclusion. After nine months of implementing the programmes in the schools, we found that there were more cooperation between student-student, teacher-student, teacher-parent, and student-parent and the school regained trust from the community. Our ideas were accepted well by the government as our director was promoted to be the Vice Minister of Education in order to implement our programmes into national education system. We consider that the key of our success is that we do practical things. We are still continuing, improving, and learning from our work with hope that the quality of Thai education will improve in near future.

Keywords: education reform, educational psychology, effective teaching, teacher-student relationship

Procedia PDF Downloads 426

Authors: Sabrine Ammar, Mohamed Tahar Bhiri

Abstract:

Automatic planning has a de facto standard language called Planning Domain Definition Language (PDDL) for describing planning problems. It aims to formalize the planning problems described by the concept of state space. PDDL-related dynamic analysis tools, namely planners and validators, are insufficient for verifying and validating PDDL descriptions. Indeed, these tools made it possible to detect errors a posteriori by means of test activity. In this paper, we recommend a formal approach coupling the two languages Event-B and PDDL, for automatic planning. Event-B is used for formal modeling by stepwise refinement with mathematical proofs of planning problems. Thus, this paper proposes a refinement strategy allowing to obtain reliable PDDL descriptions from an ultimate Event-B model correct by construction. The ultimate Event-B model, correct by construction which is supposed to be translatable into PDDL, is automatically translated into PDDL using our MDE Event-B2PDDL tool.

Keywords: code generation, event-b, PDDL, refinement strategy, translation rules

Procedia PDF Downloads 188

Authors: Saurabh Sharma, Yong Sun, Bruce Vernham

Abstract:

Accurate prediction of NOx emission is a continuous challenge in the field of diesel engine-out emission modeling. Performing experiments for each conditions and scenario cost significant amount of money and man hours, therefore model-based development strategy has been implemented in order to solve that issue. NOx formation is highly dependent on the burn gas temperature and the O₂ concentration inside the cylinder. The current empirical models are developed by calibrating the parameters representing the engine operating conditions with respect to the measured NOx. This makes the prediction of purely empirical models limited to the region where it has been calibrated. An alternative solution to that is presented in this paper, which focus on the utilization of in-cylinder combustion parameters to form a predictive semi-empirical NOx model. The result of this work is shown by developing a fast and predictive NOx model by using the physical parameters and empirical correlation. The model is developed based on the steady state data collected at entire operating region of the engine and the predictive combustion model, which is developed in Gamma Technology (GT)-Power by using Direct Injected (DI)-Pulse combustion object. In this approach, temperature in both burned and unburnt zone is considered during the combustion period i.e. from Intake Valve Closing (IVC) to Exhaust Valve Opening (EVO). Also, the oxygen concentration consumed in burnt zone and trapped fuel mass is also considered while developing the reported model.  Several statistical methods are used to construct the model, including individual machine learning methods and ensemble machine learning methods. A detailed validation of the model on multiple diesel engines is reported in this work. Substantial numbers of cases are tested for different engine configurations over a large span of speed and load points. Different sweeps of operating conditions such as Exhaust Gas Recirculation (EGR), injection timing and Variable Valve Timing (VVT) are also considered for the validation. Model shows a very good predictability and robustness at both sea level and altitude condition with different ambient conditions. The various advantages such as high accuracy and robustness at different operating conditions, low computational time and lower number of data points requires for the calibration establishes the platform where the model-based approach can be used for the engine calibration and development process. Moreover, the focus of this work is towards establishing a framework for the future model development for other various targets such as soot, Combustion Noise Level (CNL), NO₂/NOx ratio etc.

Keywords: diesel engine, machine learning, NOₓ emission, semi-empirical

Procedia PDF Downloads 110

Authors: Dario Barsi, Marina Ubaldi, Pietro Zunino, Robert Fink

Abstract:

A design and optimization procedure is proposed and developed to provide the geometry of a high efficiency compact hydraulic propeller turbine for low head. For the preliminary design of the machine, classic design criteria, based on the use of statistical correlations for the definition of the fundamental geometric parameters and the blade shapes are used. These relationships are based on the fundamental design parameters (i.e., specific speed, flow coefficient, work coefficient) in order to provide a simple yet reliable procedure. Particular attention is paid, since from the initial steps, on the correct conformation of the meridional channel and on the correct arrangement of the blade rows. The preliminary geometry thus obtained is used as a starting point for the hydrodynamic optimization procedure, carried out using a CFD calculation software coupled with a genetic algorithm that generates and updates a large database of turbine geometries. The optimization process is performed using a commercial approach that solves the turbulent Navier Stokes equations (RANS) by exploiting the axial-symmetric geometry of the machine. The geometries generated within the database are therefore calculated in order to determine the corresponding overall performance. In order to speed up the optimization calculation, an artificial neural network (ANN) based on the use of an objective function is employed. The procedure was applied for the specific case of a propeller turbine with an innovative design of a modular type, specific for applications characterized by very low heads. The procedure is tested in order to verify its validity and the ability to automatically obtain the targeted net head and the maximum for the total to total internal efficiency.

Keywords: renewable energy conversion, hydraulic turbines, low head hydraulic energy, optimization design

Procedia PDF Downloads 146

Authors: Mohammad H. Taheri, David J. Brown, Nasser Sherkat

Abstract:

Engagement is one of the most important factors in determining successful outcomes and deep learning in students. Existing approaches to detect student engagement involve periodic human observations that are subject to inter-rater reliability. Our solution uses real-time multimodal multisensor data labeled by objective performance outcomes to infer the engagement of students. The study involves four students with a combined diagnosis of cerebral palsy and a learning disability who took part in a 3-month trial over 59 sessions. Multimodal multisensor data were collected while they participated in a continuous performance test. Eye gaze, electroencephalogram, body pose, and interaction data were used to create a model of student engagement through objective labeling from the continuous performance test outcomes. In order to achieve this, a type of continuous performance test is introduced, the Seek-X type. Nine features were extracted including high-level handpicked compound features. Using leave-one-out cross-validation, a series of different machine learning approaches were evaluated. Overall, the random forest classification approach achieved the best classification results. Using random forest, 93.3% classification for engagement and 42.9% accuracy for disengagement were achieved. We compared these results to outcomes from different models: AdaBoost, decision tree, k-Nearest Neighbor, naïve Bayes, neural network, and support vector machine. We showed that using a multisensor approach achieved higher accuracy than using features from any reduced set of sensors. We found that using high-level handpicked features can improve the classification accuracy in every sensor mode. Our approach is robust to both sensor fallout and occlusions. The single most important sensor feature to the classification of engagement and distraction was shown to be eye gaze. It has been shown that we can accurately predict the level of engagement of students with learning disabilities in a real-time approach that is not subject to inter-rater reliability, human observation or reliant on a single mode of sensor input. This will help teachers design interventions for a heterogeneous group of students, where teachers cannot possibly attend to each of their individual needs. Our approach can be used to identify those with the greatest learning challenges so that all students are supported to reach their full potential.

Keywords: affective computing in education, affect detection, continuous performance test, engagement, flow, HCI, interaction, learning disabilities, machine learning, multimodal, multisensor, physiological sensors, student engagement

Procedia PDF Downloads 86

Authors: Minghui Zhang, Sirine Fkaier, Sabri Fernana, Svetlana Kiseleva, Denis Kiselev

Abstract:

The real-time identification of bacteria and fungi is difficult because they emit much weaker signals than pollen. In 2020, Plair developed Rapid-E+, which extends abilities of Rapid-E to detect smaller bioaerosols such as bacteria and fungal spores with diameters down to 0.3 µm, while keeping the similar or even better capability for measurements of large bioaerosols like pollen. Rapid-E+ enables simultaneous measurements of (1) time-resolved, polarization and angle dependent Mie scattering patterns, (2) fluorescence spectra resolved in 16 channels, and (3) fluorescence lifetime of individual particles. Moreover, (4) it provides 2D Mie scattering images which give the full information on particle morphology. The parameters of every single bioaerosol aspired into the instrument are subsequently analysed by machine learning. Firstly, pure species of microbes, e.g., Bacillus subtilis (a species of bacteria), and Penicillium chrysogenum (a species of fungal spores), were aerosolized in a bioaerosol chamber for Rapid-E+ training. Afterwards, we tested microbes under different concentrations. We used several steps of data analysis to classify and identify microbes. All single particles were analysed by the parameters of light scattering and fluorescence in the following steps. (1) They were treated with a smart filter block to get rid of non-microbes. (2) By classification algorithm, we verified the filtered particles were microbes based on the calibration data. (3) The probability threshold (defined by the user) step provides the probability of being microbes ranging from 0 to 100%. We demonstrate how Rapid-E+ identified simultaneously microbes based on the results of Bacillus subtilis (bacteria) and Penicillium chrysogenum (fungal spores). By using machine learning, Rapid-E+ achieved identification precision of 99% against the background. The further classification suggests the precision of 87% and 89% for Bacillus subtilis and Penicillium chrysogenum, respectively. The developed algorithm was subsequently used to evaluate the performance of microbe classification and quantification in real-time. The bacteria and fungi were aerosolized again in the chamber with different concentrations. Rapid-E+ can classify different types of microbes and then quantify them in real-time. Rapid-E+ enables classifying different types of microbes and quantifying them in real-time. Rapid-E+ can identify pollen down to species with similar or even better performance than the previous version (Rapid-E). Therefore, Rapid-E+ is an all-in-one instrument which classifies and quantifies not only pollen, but also bacteria and fungi. Based on the machine learning platform, the user can further develop proprietary algorithms for specific microbes (e.g., virus aerosols) and other aerosols (e.g., combustion-related particles that contain polycyclic aromatic hydrocarbons).

Keywords: bioaerosols, laser-induced fluorescence, Mie-scattering, microorganisms

Procedia PDF Downloads 87

Authors: Faiza Belbachir, Thibault Schienhinski

Abstract:

The massive development of online social networks allows users to post and share their opinions on various topics. With this huge volume of opinion, it is interesting to extract and interpret these information for different domains, e.g., product and service benchmarking, politic, system of recommendation. This is why opinion detection is one of the most important research tasks. It consists on differentiating between opinion data and factual data. The difficulty of this task is to determine an approach which returns opinionated document. Generally, there are two approaches used for opinion detection i.e. Lexical based approaches and Machine Learning based approaches. In Lexical based approaches, a dictionary of sentimental words is used, words are associated with weights. The opinion score of document is derived by the occurrence of words from this dictionary. In Machine learning approaches, usually a classifier is trained using a set of annotated document containing sentiment, and features such as n-grams of words, part-of-speech tags, and logical forms. Majority of these works are based on documents text to determine opinion score but dont take into account if these texts are really correct. Thus, it is interesting to exploit other information to improve opinion detection. In our work, we will develop a new way to consider the opinion score. We introduce the notion of trust score. We determine opinionated documents but also if these opinions are really trustable information in relation with topics. For that we use lexical SentiWordNet to calculate opinion and trust scores, we compute different features about users like (numbers of their comments, numbers of their useful comments, Average useful review). After that, we combine opinion score and trust score to obtain a final score. We applied our method to detect trust opinions in TRIPADVISOR collection. Our experimental results report that the combination between opinion score and trust score improves opinion detection.

Keywords: Tripadvisor, opinion detection, SentiWordNet, trust score

Procedia PDF Downloads 196

Authors: Nico Rosamilia

Abstract:

The study evaluates the importance of non-financial disclosure practices for regulators, investors, businesses, and markets. It aims to create a sector-specific set of indicators for environmental, social, and governance (ESG) performances alternative to the ratings of the agencies. The existing literature extensively studies the implementation of ESG rating systems. Conversely, this study has a twofold outcome. Firstly, it should generalize incentive systems and governance policies for ESG and sustainable principles. Therefore, it should contribute to the EU Sustainable Finance Disclosure Regulation. Secondly, it concerns the market and the investors by highlighting successful sustainable investing. Indeed, the study contemplates the effect of ESG adoption practices on corporate value. The research explores the asset pricing angle in order to shed light on the fragmented argument on the finance of ESG. Investors may be misguided about the positive or negative effects of ESG on performances. The paper proposes a different method to evaluate ESG performances. By comparing the results of a traditional econometric approach (Lasso) with a machine learning algorithm (Random Forest), the study establishes a set of indicators for ESG performance. Therefore, the research also empirically contributes to the theoretical strands of literature regarding model selection and variable importance in a finance framework. The algorithms will spit out sector-specific indicators. This set of indicators defines an alternative to the compounded scores of ESG rating agencies and avoids the possible offsetting effect of scores. With this approach, the paper defines a sector-specific set of indicators to standardize ESG disclosure. Additionally, it tries to shed light on the absence of a clear understanding of the direction of the ESG effect on corporate value (the problem of endogeneity).

Keywords: ESG ratings, non-financial information, value of firms, sustainable finance

Procedia PDF Downloads 80

Authors: Ezayra Dubria, Leonora Yambao

Abstract:

Mother Tongue–Based Multilingual Education (MTB-MLE) is a salient part of the recent reform in the country’s Education system which is the implementation of the K to 12 Basic Education Program. Its importance is highlighted by the passing of Republic Act 10523, otherwise known as the ‘Enhanced Basic Education Act of 2013’. However, teachers, especially new teachers encounter problems in using mother tongue as medium of instruction. Fortunately, teachers are able to create strategies which address these problems. Specifically, this paper gathered the viewpoints of teachers in using mother tongue and analyzed the different problems and strategies used. The problems encountered by teachers are lack of instructional materials written in mother tongue, especially books, lack of vocabulary, lack of teacher training, and influences of social media to learners. The strategies which address these problems are translation of literary pieces and other instructional materials, vocabulary enrichment through the use of word-of-the-day and picture-word association, remedial class, storytelling, differentiated instruction, explicit teaching, individual and group activities, and utilization of multilingual teaching.

Keywords: mother tongue-based instruction, multilingualism, problems, strategies

Procedia PDF Downloads 290

Authors: Lemoussi Somia, Ouali Mohammed, Zemirline Adel

Abstract:

In precision engineering, including precision positioning, micro-manipulation, robotic systems... a majority of these applications actuated by piezo stack used the compliant amplifier mechanism to amplifying motion and guiding it as needed utilize the flexibility of their components, in this paper, we present a novel approach introducing a symmetric structure comprising three stages, featuring rectangular flexure hinges with a compact size of 77mm×42mm×10mm. This design provides the capability for rotation, translation or a combination of both movements in both directions. The system allows for a displacement of 2107.5 μm when the input displacement of PZT is 50 μm while considering the material constraints of the aluminum alloy (7075 T6) which has a maximum admissible stress of 500 MPa However, our proposed design imposes additional constraints to ensure the stress remains below 361 MPa for optimal performance. These findings were obtained through finite element simulations conducted using ANSYS Workbench. Furthermore, our module facilitates precise control of various components within robotic systems, allowing for adjustable speeds based on specific requirements or desired outcomes.

Keywords: robotic, piezoelectric, compliant mechanism, flexure hinge

Procedia PDF Downloads 77