Search results for: generalization

Authors: Zhijie Li, Wenhui Peng, Zelong Yuan, Jianchun Wang

Abstract:

Turbulence is a complex phenomenon that plays a crucial role in various fields, such as engineering, atmospheric science, and fluid dynamics. Predicting and understanding its behavior over long time scales have been challenging tasks. Traditional methods, such as large-eddy simulation (LES), have provided valuable insights but are computationally expensive. In the past few years, machine learning methods have experienced rapid development, leading to significant improvements in computational speed. However, ensuring stable and accurate long-term predictions remains a challenging task for these methods. In this study, we introduce the implicit U-net enhanced Fourier neural operator (IU-FNO) as a solution for stable and efficient long-term predictions of the nonlinear dynamics in three-dimensional (3D) turbulence. The IU-FNO model combines implicit re-current Fourier layers to deepen the network and incorporates the U-Net architecture to accurately capture small-scale flow structures. We evaluate the performance of the IU-FNO model through extensive large-eddy simulations of three types of 3D turbulence: forced homogeneous isotropic turbulence (HIT), temporally evolving turbulent mixing layer, and decaying homogeneous isotropic turbulence. The results demonstrate that the IU-FNO model outperforms other FNO-based models, including vanilla FNO, implicit FNO (IFNO), and U-net enhanced FNO (U-FNO), as well as the dynamic Smagorinsky model (DSM), in predicting various turbulence statistics. Specifically, the IU-FNO model exhibits improved accuracy in predicting the velocity spectrum, probability density functions (PDFs) of vorticity and velocity increments, and instantaneous spatial structures of the flow field. Furthermore, the IU-FNO model addresses the stability issues encountered in long-term predictions, which were limitations of previous FNO models. In addition to its superior performance, the IU-FNO model offers faster computational speed compared to traditional large-eddy simulations using the DSM model. It also demonstrates generalization capabilities to higher Taylor-Reynolds numbers and unseen flow regimes, such as decaying turbulence. Overall, the IU-FNO model presents a promising approach for long-term dynamics prediction in 3D turbulence, providing improved accuracy, stability, and computational efficiency compared to existing methods.

Keywords: data-driven, Fourier neural operator, large eddy simulation, fluid dynamics

Procedia PDF Downloads 43

Authors: Yuliya Huseva

Abstract:

To investigate the contents of tears to determine latent blood. Methods: Tear samples from 72 women were studied with the microscopy of tears aspirated with a capillary and stained by Nocht and with a chemical method of test strips with chromogen. Statistical data processing was carried out using statistical packages Statistica 10.0 for Windows, calculation of Pearson's chi-square test, Yule association coefficient, the method of determining sensitivity and specificity. Results:, In 30.6% (22) of tear samples erythrocytes were revealed microscopically. Correlations between the presence of erythrocytes in the tear and the phase of the menstrual cycle has been discovered. In the follicular phase of the cycle, erythrocytes were found in 59.1% (13) people, which is significantly more (x2=4.2, p=0.041) compared to the luteal phase - in 40.9% (9) women. In the first seven days of the follicular phase of the menstrual cycle the erythrocytes were predominanted of in the tears of women examined testifies in favour of the vicarious bleeding from the mucous membranes of extragenital organs in sync with menstruation. Of the other cellular elements in tear samples with latent haemolacria, neutrophils prevailed - in 45.5% (10), while lymphocytes were less common - in 27.3% (6), because neutrophil exudation is accompanied by vasodilatation of the conjunctiva and the release of erythrocytes into the conjunctival cavity. It was found that the prognostic significance of the chemical method was 0.53 of the microscopic method. In contrast to microscopy, which detected blood in tear samples from 30.6% (22) of women, blood was detected chemically in tears of 16.7% (12). An association between latent haemolacria and endometriosis was found (k=0.75, p≤0.05). Microscopically, in the tears of patients with endometriosis, erythrocytes were detected in 70% of cases, while in healthy women without endometriosis - in 25% of cases. The proportion of women with erythrocytes in tears, determined by a chemical method, was 41.7% among patients with endometriosis, which is significantly more (x2=6.5, p=0.011) than 11.7% among women without endometriosis. The data obtained can be explained by the etiopathogenesis of the extragenital endometriosis which is caused by hematogenous spread of endometrial tissue into the orbit. In endometriosis, erythrocytes are found against the background of accumulations of epithelial cells. In the tear samples of 4 women with endometriosis, glandular cuboidal epithelial cells, morphologically similar to endometrial cells, were found, which may indicate a generalization of the disease. Conclusions: Single erythrocytes can normally be found in the tears, their number depends on the phase of the menstrual cycle, increasing in the follicular phase. Erythrocytes found in tears against the background of accumulations of epitheliocytes and their glandular atypia may indicate a manifestation of extragenital endometriosis. Both used methods (microscopic and chemical) are informative in revealing latent haemolacria. The microscopic method is more sensitive, reveals intact erythrocytes, and besides, it provides information about other cells. At the same time, the chemical method is faster and technically simpler, it determines the presence of haemoglobin and its metabolic products, and can be used as a screening.

Keywords: tear, blood, microscopy, epitheliocytes

Procedia PDF Downloads 92

Authors: Chunhua Liao, Jinfei Wang, Bo Shan, Yang Song, Yongjun He, Taifeng Dong

Abstract:

Winter wheat is one of the main crops in Canada. Forecasting of within-field variability of yield in winter wheat at the early stages is essential for precision farming. However, the crop yield modelling based on high spatial resolution satellite data is generally affected by the lack of continuous satellite observations, resulting in reducing the generalization ability of the models and increasing the difficulty of crop yield forecasting at the early stages. In this study, the correlations between Sentinel-2 data (vegetation indices and reflectance) and yield data collected by combine harvester were investigated and a generalized multivariate linear regression (MLR) model was built and tested with data acquired in different years. It was found that the four-band reflectance (blue, green, red, near-infrared) performed better than their vegetation indices (NDVI, EVI, WDRVI and OSAVI) in wheat yield prediction. The optimum phenological stage for wheat yield prediction with highest accuracy was at the growing stages from the end of the flowering to the beginning of the filling stage. The best MLR model was therefore built to predict wheat yield before harvest using Sentinel-2 data acquired at the end of the flowering stage. Further, to improve the ability of the yield prediction at the early stages, three simple unsupervised domain adaptation (DA) methods were adopted to transform the reflectance data at the early stages to the optimum phenological stage. The winter wheat yield prediction using multiple vegetation indices showed higher accuracy than using single vegetation index. The optimum stage for winter wheat yield forecasting varied with different fields when using vegetation indices, while it was consistent when using multispectral reflectance and the optimum stage for winter wheat yield prediction was at the end of flowering stage. The average testing RMSE of the MLR model at the end of the flowering stage was 604.48 kg/ha. Near the booting stage, the average testing RMSE of yield prediction using the best MLR was reduced to 799.18 kg/ha when applying the mean matching domain adaptation approach to transform the data to the target domain (at the end of the flowering) compared to that using the original data based on the models developed at the booting stage directly (“MLR at the early stage”) (RMSE =1140.64 kg/ha). This study demonstrated that the simple mean matching (MM) performed better than other DA methods and it was found that “DA then MLR at the optimum stage” performed better than “MLR directly at the early stages” for winter wheat yield forecasting at the early stages. The results indicated that the DA had a great potential in near real-time crop yield forecasting at the early stages. This study indicated that the simple domain adaptation methods had a great potential in crop yield prediction at the early stages using remote sensing data.

Keywords: wheat yield prediction, domain adaptation, Sentinel-2, within-field scale

Procedia PDF Downloads 37

Authors: Farid Khosravikia, Patricia Clayton

Abstract:

The main objective of this study is to evaluate the advantages and disadvantages of various machine learning techniques in forecasting ground-motion intensity measures given source characteristics, source-to-site distance, and local site condition. Intensity measures such as peak ground acceleration and velocity (PGA and PGV, respectively) as well as 5% damped elastic pseudospectral accelerations at different periods (PSA), are indicators of the strength of shaking at the ground surface. Estimating these variables for future earthquake events is a key step in seismic hazard assessment and potentially subsequent risk assessment of different types of structures. Typically, linear regression-based models, with pre-defined equations and coefficients, are used in ground motion prediction. However, due to the restrictions of the linear regression methods, such models may not capture more complex nonlinear behaviors that exist in the data. Thus, this study comparatively investigates potential benefits from employing other machine learning techniques as a statistical method in ground motion prediction such as Artificial Neural Network, Random Forest, and Support Vector Machine. The algorithms are adjusted to quantify event-to-event and site-to-site variability of the ground motions by implementing them as random effects in the proposed models to reduce the aleatory uncertainty. All the algorithms are trained using a selected database of 4,528 ground-motions, including 376 seismic events with magnitude 3 to 5.8, recorded over the hypocentral distance range of 4 to 500 km in Oklahoma, Kansas, and Texas since 2005. The main reason of the considered database stems from the recent increase in the seismicity rate of these states attributed to petroleum production and wastewater disposal activities, which necessities further investigation in the ground motion models developed for these states. Accuracy of the models in predicting intensity measures, generalization capability of the models for future data, as well as usability of the models are discussed in the evaluation process. The results indicate the algorithms satisfy some physically sound characteristics such as magnitude scaling distance dependency without requiring pre-defined equations or coefficients. Moreover, it is shown that, when sufficient data is available, all the alternative algorithms tend to provide more accurate estimates compared to the conventional linear regression-based method, and particularly, Random Forest outperforms the other algorithms. However, the conventional method is a better tool when limited data is available.

Keywords: artificial neural network, ground-motion models, machine learning, random forest, support vector machine

Procedia PDF Downloads 92

Authors: Maggie Mosher, Adam Carreon, Sean Smith

Abstract:

A systematic literature review was conducted to explore the evidence base on the use of augmented reality (AR), virtual reality (VR), mixed reality (MR), and extended reality (XR) to present social skill instruction to school-age students with autism spectrum disorder (ASD). Specifically, the systematic review focus was on a. the participants and intervention agents using AR, VR, MR, and XR for social skill acquisition b. the social skills taught through these mediums and c. the social validity measures (i.e., goals, procedures, and outcomes) reported in these studies. Forty-one articles met the inclusion criteria. Researchers in six studies taught social skills to students through AR, in 27 studies through non-immersive VR, and in 10 studies through immersive VR. No studies used MR or XR. The primary targeted social skills were relationship skills, emotion recognition, social awareness, cooperation, and executive functioning. An intervention to improve many social skills was implemented by 73% of researchers, 17% taught a single skill, and 10% did not clearly state the targeted skill. The intervention was considered effective in 26 of the 41 studies (63%), not effective in four studies (10%), and 11 studies (27%) reported mixed results. No researchers reported information for all 17 social validity indicators. The social validity indicators reported by researchers ranged from two to 14. Social validity measures on the feelings toward and use of the technology were provided in 22 studies (54%). Findings indicated both AR and VR are promising platforms for providing social skill instruction to students with ASD. Studies utilizing this technology show a number of social validity indicators. However, the limited information provided on the various interventions, participant characteristics, and validity measures, offers insufficient evidence of the impact of these technologies in teaching social skills to students with ASD. Future research should develop a protocol for training treatment agents to assess the role of different variables (i.e., whether agents are customizing content, monitoring student learning, using intervention specific vocabulary in their day to day instruction). Sustainability may be increased by providing training in the technology to both treatment agents and participants. Providing scripts of instruction occurring within the intervention would provide the needed information to determine the primary method of teaching within the intervention. These variables play a role in maintenance and generalization of the social skills. Understanding the type of feedback provided would help researchers determine if students were able to feel rewarded for progressing through the scenarios or if students require rewarding aspects within the intervention (i.e., badges, trophies). AR has the potential to generalize instruction and VR has the potential for providing a practice environment for performance deficits. Combining these two technologies into a mixed reality intervention may provide a more cohesive and effective intervention.

Keywords: autism, augmented reality, social and emotional learning, social skills, virtual reality

Procedia PDF Downloads 82

Authors: Rafael do Amaral Teodoro, Leandro Augusto da Silva

Abstract:

Originated from graphite, graphene is a two-dimensional (2D) material that promises to revolutionize technology in many different areas, such as energy, telecommunications, civil construction, aviation, textile, and medicine. This is possible because its structure, formed by carbon bonds, provides desirable optical, thermal, and mechanical characteristics that are interesting to multiple areas of the market. Thus, several research and development centers are studying different manufacturing methods and material applications of graphene, which are often compromised by the scarcity of more agile and accurate methodologies to characterize the material – that is to determine its composition, shape, size, and the number of layers and crystals. To engage in this search, this study proposes a computational methodology that applies deep learning to identify graphene oxide crystals in order to characterize samples by crystal sizes. To achieve this, a fully convolutional neural network called U-net has been trained to segment SEM graphene oxide images. The segmentation generated by the U-net is fine-tuned with a standard deviation technique by classes, which allows crystals to be distinguished with different labels through an object delimitation algorithm. As a next step, the characteristics of the position, area, perimeter, and lateral measures of each detected crystal are extracted from the images. This information generates a database with the dimensions of the crystals that compose the samples. Finally, graphs are automatically created showing the frequency distributions by area size and perimeter of the crystals. This methodological process resulted in a high capacity of segmentation of graphene oxide crystals, presenting accuracy and F-score equal to 95% and 94%, respectively, over the test set. Such performance demonstrates a high generalization capacity of the method in crystal segmentation, since its performance considers significant changes in image extraction quality. The measurement of non-overlapping crystals presented an average error of 6% for the different measurement metrics, thus suggesting that the model provides a high-performance measurement for non-overlapping segmentations. For overlapping crystals, however, a limitation of the model was identified. To overcome this limitation, it is important to ensure that the samples to be analyzed are properly prepared. This will minimize crystal overlap in the SEM image acquisition and guarantee a lower error in the measurements without greater efforts for data handling. All in all, the method developed is a time optimizer with a high measurement value, considering that it is capable of measuring hundreds of graphene oxide crystals in seconds, saving weeks of manual work.

Keywords: characterization, graphene oxide, nanomaterials, U-net, deep learning

Procedia PDF Downloads 134

Authors: Festa Shabani

Abstract:

This paper aims at investigating bilingual speech in the International Schools of Prishtina. More particularly, it seeks to analyze bilingual phenomena among adolescent students highly exposed to English with the latter as the language of instruction at school in naturally-occurring conversations within school environment. Adolescence was deliberately chosen since it is regarded as an age when peer influence on language choice is the greatest. Driven by daily unsystematic observation and prior research already undertaken, the hypothesis stated is that Albanian continues to be the dominant language among Prishtina international schools’ students with a lot of code-switched items from the English. Furthermore, they will also use lexical borrowings - words already adapted in the receiving language, from the language they have been in contact with, in their speech often in the lack of existing equivalents in Albanian or for other reasons. This is done owing to the fact that the language of instruction at school is English, and any topic related to the language they have been exposed to will trigger them to use English. Therefore, this needs special attention in an attempt to identify patterns of their speech; in this way, linguistic and socio-pragmatic factors will be considered when analyzing the motivations behind their language choice. Methodology for collecting data include participant systematic observation and tape-recording. While observing them in their natural conversations, the fieldworker also took notes, which helped transcribe details better. The paper starts by raising the question of whether code-switching is occurring among Prishtina International Schools’ students highly exposed to English. The data gathered from students in informal settings suggests that there are well-founded grounds for an affirmative answer. The participants in this study are observed to be code-switching, although showing differences in degree. However, a generalization cannot be made on the basis of the findings except in so far it appears that English has, in turn, became a language to which they turn when identifying with the group when discussing about particular school topics. Particularly, participants seemed to use intra-sentential CS in cases when they seem to find an English expression rather easier than an Albanian one when repeating or emphasizing a point when urged to talk about educational issues with English being their language of instruction, and inter-sentential code-switching, particularly when quoting others. Concerning the grammatical aspect of code-switching, the intrasentential CS is used more than the intersentetial one. Speaking of gender, the results show that there were really no significant differences in regards quantity between male and female participants. However, the slight tendency for men to code switch intrasententially more than women was manifested. Similarly, a slight tendency again for a difference to emerge is on intersentential switching, which contributes 21% to the total number of switches for women, but 11% to the total number of switches for men.

Keywords: Albanian, code-switching contact linguistics, bilingual phenomena, lexical borrowing, English

Procedia PDF Downloads 97

Authors: Venkateswarlu Kappara

Abstract:

The term ‘Dharma’ is complex and ubiquitous. It has no equivalent word in English Initially applied to Aryans. In Rig Veda, it appears in a number of places with different meanings. The word Dharma comes from the roots word ‘dhr’ (Dhri-Dharayatetiiti Dharmaha). Principles of Dharma are all pervading. The closest synonyms for Dharma in English is ‘Righteousness.’ In a holy book Mahabharata, it is mentioned that Dharma destroys those who destroy it, Dharma Protects those who protect it. Also, Dharma might be shadowed, now and then by evil forces, but at the end, Dharma always triumphs. This line embodies the eternal victory of good over evil. In Mahabharata, Lord Krishna says Dharma upholds both, this worldly and other worldly affairs. Rig Veda says, ‘O Indra! Lead us on the path of Rta, on the right path over all evils.’ For Buddhists, Dharma most often means the body of teachings expounded by the Buddha. The Dharma is one of the three Jewels (Tri Ratnas) of Buddhism under which the followers take refuge. They are: the ‘Buddha’ meaning the minds perfection or enlightenment, the Dharma, meaning the teachings and the methods of the Buddha, and the Sangha meaning those awakened people who provide guidance and support followers. Buddha denies a separate permanent ‘I.’ Buddha Accepts Suffering (Dukka). Change / impermanence (Anicca) and not– self (Annatta) Dharma in the Buddhist scriptures has a variety of meanings including ‘phenomenon’ and ‘nature’ or ‘characteristic.’ For Sikhs, the word ‘Dharma’ means the ‘path’ of righteousness’ The Sikh scriptures attempt to answer the exposition of Dharma. The main Holy Scripture of the Sikh religion is called the Guru Granth Sahib. The faithful people are fully bound to do whatever the Dharma wants them to do. Such is the name of the Immaculate Lord. Only one who has faith comes to know such a state of mind. The righteous judge of Dharma, by the Hukam of God’s Command, sits and Administers true justice. From Dharma flow wealth and pleasure. The study indicates that in Sikh religion, the Dharma is the path of righteousness; In Buddhism, the mind’s perfection of enlightenment, and in Hinduism, it is non-violence, purity, truth, control of senses, not coveting the property of others. The comparative study implies that all religions dealt with Dharma for welfare of the mankind. The methodology adapted is theoretical, analytical and comparative. The present study indicates how far Indian philosophical systems influenced the present circumstances and how far the present system is not compatible with Ancient philosophical systems. A tentative generalization would be that the present system which is mostly influenced by the British Governance may not totally reflect the ancient norms. However, the mental make-up continues to be influenced by Ancient philosophical systems.

Keywords: Dharma, Dukka (suffering), Rakshati, righteous

Procedia PDF Downloads 140

Authors: Jenna Pyorala

Abstract:

This research looks at the principles and purposes ornamental design has served in the field of textile design. Ornamental designs are characterized by richness of details, abundance of elements, vegetative motifs and organic forms that flow harmoniously in complex compositions. Research on ornamental design is significant, because ornaments have been overlooked and considered as less meaningful and aesthetically pleasing than minimalistic, modern designs. This is despite the fact that in many parts of the world ornaments have been an important part of the cultural identification and expression for centuries. Ornament has been claimed to be superficial and merely used as a decorative way to hide the faults of designs. Such generalization is an incorrect interpretation of the real purposes of ornament. Many ornamental patterns tell stories, present mythological scenes or convey symbolistic meanings. Historically, ornamental decorations have been representing ideas and characteristics such as abundance, wealth, power and personal magnificence. The production of fine ornaments required refined skill, eye for intricate detail and perseverance while compiling complex elements into harmonious compositions. For this reason, ornaments have played an important role in the advancement of craftsmanship. Even though it has been claimed that people in the western design world have lost the relationship to ornament, the relation to it has merely changed from the practice of a craftsman to conceptualisation of a designer. With the help of new technological tools the production of ornaments has become faster and more efficient, demanding less manual labour. Designers who commit to this style of organic forms and vegetative motifs embrace and respect nature by representing its organically growing forms and by following its principles. The complexity of the designs is used as a way to evoke a sense of extraordinary beauty and stimulate intellect by freeing the mind from the predetermined interpretations. Through the study of these purposes it can be demonstrated that complex and richer design styles are as valuable a part of the world of design as more modern design approaches. The study highlights the meaning of ornaments by presenting visual examples and literature research findings. The practice based part of the project is the visual analysis of historical and cultural ornamental traditions such as Indian Chikan embroidery, Persian carpets, Art Nouveau and Rococo according to the rubric created for the purpose. The next step is the creation of ornamental designs based on the key elements in different styles. Theoretical and practical parts are woven together in this study that respects respect the long traditions of ornaments and highlight the importance of these design approaches to the field, in contrast to the more commonly preferred styles.

Keywords: cultural design traditions, ornamental design, organic forms from nature, textile design

Procedia PDF Downloads 204

Authors: M. Venegas, M. De Vega, N. García-Hernando

Abstract:

Absorption cooling chillers have received growing attention over the past few decades as they allow the use of low-grade heat to produce the cooling effect. The combination of this technology with solar thermal energy in the summer period can reduce the electricity consumption peak due to air-conditioning. One of the main components, the absorber, is designed for simultaneous heat and mass transfer. Usually, shell and tubes heat exchangers are used, which are large and heavy. Cooling water from a cooling tower is conventionally used to extract the heat released during the absorption and condensation processes. These are clear inconvenient for the generalization of the absorption technology use, limiting its benefits in the contribution to the reduction in CO2 emissions, particularly for the H2O-LiBr solution which can work with low heat temperature sources as provided by solar panels. In the present work a promising new technology is under study, consisting in the use of membrane contactors in adiabatic microchannel mass exchangers. The configuration here proposed consists in one or several modules (depending on the cooling capacity of the chiller) that contain two vapour channels, separated from the solution by adjacent microporous membranes. The solution is confined in rectangular microchannels. A plastic or synthetic wall separates the solution channels between them. The solution entering the absorber is previously subcooled using ambient air. In this way, the need for a cooling tower is avoided. A model of the configuration proposed is developed based on mass and energy balances and some correlations were selected to predict the heat and mass transfer coefficients. The concentration and temperatures along the channels cannot be explicitly determined from the set of equations obtained. For this reason, the equations were implemented in a computer code using Engineering Equation Solver software, EES™. With the aim of minimizing the absorber volume to reduce the size of absorption cooling chillers, the ratio between the cooling power of the chiller and the absorber volume (R) is calculated. Its variation is shown along the solution channels, allowing its optimization for selected operating conditions. For the case considered the solution channel length is recommended to be lower than 3 cm. Maximum values of R obtained in this work are higher than the ones found in optimized horizontal falling film absorbers using the same solution. Results obtained also show the variation of R and the chiller efficiency (COP) for different ambient temperatures and desorption temperatures typically obtained using flat plate solar collectors. The configuration proposed of adiabatic membrane-based absorber using ambient air to subcool the solution is a good technology to reduce the size of the absorption chillers, allowing the use of low temperature solar heat and avoiding the need for cooling towers.

Keywords: adiabatic absorption, air-cooled, membrane, solar thermal energy

Procedia PDF Downloads 255

Authors: Vivian Caroline Koerbel Dombrowski

Abstract:

Armed conflicts have always meant the ultimate expression of power and at the same time, lack of understanding among nations. Cities were destroyed, people were killed, assets were devastated. But these are not only the loss of a war: the environmental damage comes to be considered immeasurable losses in the short, medium and long term. And this is because no nation wants to bear that cost. They invest in military equipment, training, technical equipment but the environmental account yet finds gaps in international law. Considering such a generalization in rights protection, many nations are at imminent danger in a conflict if the water will be used as a mass weapon, especially if we consider important rivers such as Jordan, Euphrates and Nile. The top three international documents were analyzed on the subject: the Stockholm Convention (1972), Additional Protocol I to the Geneva Convention (1977) and the Rome Statute (1998). Indeed, some references are researched in doctrine, especially scientific articles, to substantiate with consistent data about the extent of the damage, historical factors and decisions which have been successful. However, due to the lack of literature about this subject, the research tends to be exhaustive. From the study of the indicated material, it was noted that international law - humanitarian and environmental - calls in some of its instruments the environmental protection in war conflicts, but they are generic and vague rules that do not define exactly what is the environmental damage , nor sets standards for measure them. Taking into account the mains conflicts of the century XX: World War II, the Vietnam War and the Gulf War, one must realize that the environmental consequences were of great rides - never deactivated landmines, buried nuclear weapons, armaments and munitions destroyed in the soil, chemical weapons, not to mention the effects of some weapons when used (uranium, agent Orange, etc). Extending the search for more recent conflicts such as Afghanistan, it is proven that the effects on health of the civilian population were catastrophic: cancer, birth defects, and deformities in newborns. There are few reports of nations that, somehow, repaired the damage caused to the environment as a result of the conflict. In the pitch of contemporary conflicts, many nations fear that water resources are used as weapons of mass destruction, because once contaminated - directly or indirectly - can become a means of disguised genocide side effect of military objective. In conclusion, it appears that the main international treaties governing the subject mention the concern for environmental protection, however leave the normative specifications vacancies necessary to effectively there is a prevention of environmental damage in armed conflict and, should they occur, the repair of the same. Still, it appears that there is no protection mechanism to safeguard natural resources and avoid them to become a mass destruction weapon.

Keywords: armed conflicts, criminal liability, environmental damages, humanitarian law, mass weapon

Procedia PDF Downloads 392

Authors: Adriana Postal, Silvio C. Sampaio, Marcio A. Villas Boas, Josué P. Castro, Ralpho R. Reis

Abstract:

This study deals with the estimation of reference evapotranspiration (ET₀) in an agricultural context, focusing on efficient irrigation management to meet the growing interest in the sustainable management of water resources. Given the importance of water in agriculture and its scarcity in many regions, efficient use of this resource is essential to ensure food security and environmental sustainability. The methodology used involved the application of artificial intelligence techniques, specifically Multilayer Perceptron (MLP) Artificial Neural Networks (ANNs), to predict ET₀ in the state of Paraná, Brazil. The models were trained and validated with meteorological data from the Brazilian National Institute of Meteorology (INMET), together with data obtained from a producer's weather station in the western region of Paraná. Two optimizers (SGD and Adam) and different meteorological variables, such as temperature, humidity, solar radiation, and wind speed, were explored as inputs to the models. Nineteen configurations with different input variables were tested; amidst them, configuration 9, with 8 input variables, was identified as the most efficient of all. Configuration 10, with 4 input variables, was considered the most effective, considering the smallest number of variables. The main conclusions of this study show that MLP ANNs are capable of accurately estimating ET₀, providing a valuable tool for irrigation management in agriculture. Both configurations (9 and 10) showed promising performance in predicting ET₀. The validation of the models with cultivator data underlined the practical relevance of these tools and confirmed their generalization ability for different field conditions. The results of the statistical metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Coefficient of Determination (R2), showed excellent agreement between the model predictions and the observed data, with MAE as low as 0.01 mm/day and 0.03 mm/day, respectively. In addition, the models achieved an R2 between 0.99 and 1, indicating a satisfactory fit to the real data. This agreement was also confirmed by the Kolmogorov-Smirnov test, which evaluates the agreement of the predictions with the statistical behavior of the real data and yields values between 0.02 and 0.04 for the producer data. In addition, the results of this study suggest that the developed technique can be applied to other locations by using specific data from these sites to further improve ET₀ predictions and thus contribute to sustainable irrigation management in different agricultural regions. To summarize, this study has helped to advance research in the field of irrigation management in agriculture. It provides an accessible and effective approach to ET₀ estimation that has the potential to significantly improve water use efficiency and promote agricultural sustainability in different contexts.

Keywords: agricultural technology, neural networks in agriculture, water efficiency, water use optimization

Procedia PDF Downloads 12

Authors: Maria Evnevich

Abstract:

In modern conditions, in most markets, price competition is becoming less effective. On the one hand, there is a gradual decrease in the level of marginality in main traditional sectors of the economy, so further price reduction becomes too ‘expensive’ for the company. On the other hand, the effect of price reduction is leveled, and the reason for this phenomenon is likely to be informational. As a result, it turns out that even if the company reduces prices, making its products more accessible to the buyer, there is a high probability that this will not lead to increase in sales unless additional large-scale advertising and information campaigns are conducted. Similarly, a large-scale information and advertising campaign have a much greater effect itself than price reductions. At the same time, the cost of mass informing is growing every year, especially when using the main information channels. The article presents generalization, systematization and development of theoretical approaches and best practices in the field of customer focus approach to business management and in the field of relationship marketing in the modern digital economy. The research methodology is based on the synthesis and content-analysis of sociological and marketing research and on the study of the systems of working with consumer appeals and loyalty programs in the 50 largest client-oriented companies in Russia. Also, the analysis of internal documentation on customers’ purchases in one of the largest retail companies in Russia allowed to identify if buyers prefer to buy goods for complex purchases in one retail store with the best price image for them. The cost of attracting a new client is now quite high and continues to grow, so it becomes more important to keep him and increase the involvement through marketing tools. A huge role is played by modern digital technologies used both in advertising (e-mailing, SEO, contextual advertising, banner advertising, SMM, etc.) and in service. To implement the above-described client-oriented omnichannel service, it is necessary to identify the client and work with personal data provided when filling in the loyalty program application form. The analysis of loyalty programs of 50 companies identified the following types of cards: discount cards, bonus cards, mixed cards, coalition loyalty cards, bank loyalty programs, aviation loyalty programs, hybrid loyalty cards, situational loyalty cards. The use of loyalty cards allows not only to stimulate the customer to purchase ‘untargeted’, but also to provide individualized offers, as well as to produce more targeted information. The development of digital technologies and modern means of communication has significantly changed not only the sphere of marketing and promotion, but also the economic landscape as a whole. Factors of competitiveness are the digital opportunities of companies in the field of customer orientation: personalization of service, customization of advertising offers, optimization of marketing activity and improvement of logistics.

Keywords: customer focus, digital economy, loyalty program, relationship marketing

Procedia PDF Downloads 136

Authors: María-Luisa Pinto-Salamanca, Wilson-Javier Pérez-Holguín

Abstract:

Reconstruction of contact forces is a fundamental technique for analyzing the properties of a touched object and is essential for regulating the grip force in slip control loops. This is based on the processing of the distribution, intensity, and direction of the forces during the capture of the sensors. Currently, efficient hardware alternatives have been used more frequently in different fields of application, allowing the implementation of computationally complex algorithms, as is the case with tactile signal processing. The use of hardware for smart tactile sensing systems is a research area that promises to improve the processing time and portability requirements of applications such as artificial skin and robotics, among others. The literature review shows that hardware implementations are present today in almost all stages of smart tactile detection systems except in the force reconstruction process, a stage in which they have been less applied. This work presents a hardware implementation of a model-driven reported in the literature for the contact force reconstruction of flat and rigid tactile sensor arrays from normal stress data. From the analysis of a software implementation of such a model, this implementation proposes the parallelization of tasks that facilitate the execution of matrix operations and a two-dimensional optimization function to obtain a vector force by each taxel in the array. This work seeks to take advantage of the parallel hardware characteristics of Field Programmable Gate Arrays, FPGAs, and the possibility of applying appropriate techniques for algorithms parallelization using as a guide the rules of generalization, efficiency, and scalability in the tactile decoding process and considering the low latency, low power consumption, and real-time execution as the main parameters of design. The results show a maximum estimation error of 32% in the tangential forces and 22% in the normal forces with respect to the simulation by the Finite Element Modeling (FEM) technique of Hertzian and non-Hertzian contact events, over sensor arrays of 10×10 taxels of different sizes. The hardware implementation was carried out on an MPSoC XCZU9EG-2FFVB1156 platform of Xilinx® that allows the reconstruction of force vectors following a scalable approach, from the information captured by means of tactile sensor arrays composed of up to 48 × 48 taxels that use various transduction technologies. The proposed implementation demonstrates a reduction in estimation time of x / 180 compared to software implementations. Despite the relatively high values of the estimation errors, the information provided by this implementation on the tangential and normal tractions and the triaxial reconstruction of forces allows to adequately reconstruct the tactile properties of the touched object, which are similar to those obtained in the software implementation and in the two FEM simulations taken as reference. Although errors could be reduced, the proposed implementation is useful for decoding contact forces for portable tactile sensing systems, thus helping to expand electronic skin applications in robotic and biomedical contexts.

Keywords: contact forces reconstruction, forces estimation, tactile sensor array, hardware implementation

Procedia PDF Downloads 150

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 259

Authors: Teresita Lopez Joy, Jose A. Acevedo Suarez, Martha I. Gomez Acosta, Ana Julia Acevedo Urquiaga

Abstract:

Current trends in business competitiveness indicate the need to manage businesses as supply chains and not in isolation. The use of strategies aimed at maximum satisfaction of customers in a network and based on inter-company cooperation; contribute to obtaining successful joint results. In the Cuban economic context, the development of productive linkages to achieve integrated management of supply chains is considering a key aspect. In order to achieve this jump, it is necessary to develop acting capabilities in the entities that make up the chains through a systematic procedure that allows arriving at a management model in consonance with the environment. The objective of the research focuses on: designing a model and procedure for the development of integrated management of supply chains in economic entities. The results obtained are: the Model and the Procedure for the Development of the Supply Chains Integrated Management (MP-SCIM). The Model is based on the development of logistics in the network actors, the joint work between companies, collaborative planning and the monitoring of a main indicator according to the end customers. The application Procedure starts from the well-founded need for development in a supply chain and focuses on training entrepreneurs as doers. The characterization and diagnosis is done to later define the design of the network and the relationships between the companies. It takes into account the feedback as a method of updating the conditions and way to focus the objectives according to the final customers. The MP-SCIM is the result of systematic work with a supply chain approach in companies that have consolidated as coordinators of their network. The cases of the edible oil chain and explosives for construction sector reflect results of more remarkable advances since they have applied this approach for more than 5 years and maintain it as a general strategy of successful development. The edible oil trading company experienced a jump in sales. In 2006, the company started the analysis in order to define the supply chain, apply diagnosis techniques, define problems and implement solutions. The involvement of the management and the progressive formation of performance capacities in the personnel allowed the application of tools according to the context. The company that coordinates the explosives chain for construction sector shows adequate training with independence and opportunity in the face of different situations and variations of their business environment. The appropriation of tools and techniques for the analysis and implementation of proposals is a characteristic feature of this case. The coordinating entity applies integrated supply chain management to its decisions based on the timely training of the necessary action capabilities for each situation. Other cases of study and application that validate these tools are also detailed in this paper, and they highlight the results of generalization in the quantitative and qualitative improvement according to the final clients. These cases are: teaching literature in universities, agricultural products of local scope and medicine supply chains.

Keywords: integrated management, logistic system, supply chain management, tactical-operative planning

Procedia PDF Downloads 122

Authors: Bitewulign Mekonnen

Abstract:

Context: This scientific paper focuses on the use of near-infrared (NIR) spectroscopy to determine glucose concentration in aqueous solutions accurately and rapidly. The study compares six different machine learning methods for predicting glucose concentration and also explores the development of a deep learning model for classifying NIR spectra. The objective is to optimize the detection model and improve the accuracy of glucose prediction. This research is important because it provides a comprehensive analysis of various machine-learning techniques for estimating aqueous glucose concentrations. Research Aim: The aim of this study is to compare and evaluate different machine-learning methods for predicting glucose concentration from NIR spectra. Additionally, the study aims to develop and assess a deep-learning model for classifying NIR spectra. Methodology: The research methodology involves the use of machine learning and deep learning techniques. Six machine learning regression models, including support vector machine regression, partial least squares regression, extra tree regression, random forest regression, extreme gradient boosting, and principal component analysis-neural network, are employed to predict glucose concentration. The NIR spectra data is randomly divided into train and test sets, and the process is repeated ten times to increase generalization ability. In addition, a convolutional neural network is developed for classifying NIR spectra. Findings: The study reveals that the SVMR, ETR, and PCA-NN models exhibit excellent performance in predicting glucose concentration, with correlation coefficients (R) > 0.99 and determination coefficients (R²)> 0.985. The deep learning model achieves high macro-averaging scores for precision, recall, and F1-measure. These findings demonstrate the effectiveness of machine learning and deep learning methods in optimizing the detection model and improving glucose prediction accuracy. Theoretical Importance: This research contributes to the field by providing a comprehensive analysis of various machine-learning techniques for estimating glucose concentrations from NIR spectra. It also explores the use of deep learning for the classification of indistinguishable NIR spectra. The findings highlight the potential of machine learning and deep learning in enhancing the prediction accuracy of glucose-relevant features. Data Collection and Analysis Procedures: The NIR spectra and corresponding references for glucose concentration are measured in increments of 20 mg/dl. The data is randomly divided into train and test sets, and the models are evaluated using regression analysis and classification metrics. The performance of each model is assessed based on correlation coefficients, determination coefficients, precision, recall, and F1-measure. Question Addressed: The study addresses the question of whether machine learning and deep learning methods can optimize the detection model and improve the accuracy of glucose prediction from NIR spectra. Conclusion: The research demonstrates that machine learning and deep learning methods can effectively predict glucose concentration from NIR spectra. The SVMR, ETR, and PCA-NN models exhibit superior performance, while the deep learning model achieves high classification scores. These findings suggest that machine learning and deep learning techniques can be used to improve the prediction accuracy of glucose-relevant features. Further research is needed to explore their clinical utility in analyzing complex matrices, such as blood glucose levels.

Keywords: machine learning, signal processing, near-infrared spectroscopy, support vector machine, neural network

Procedia PDF Downloads 56

Authors: Alon Cohen, Jeffrey Kantor, Shalom Levy

Abstract:

The use of the audit risk model in auditing has faced limitations and difficulties, leading auditors to rely on a conceptual level of its application. The qualitative approach to assessing risks has resulted in different risk assessments, affecting the quality of audits and decision-making on the adoption of CAATTs. This study aims to investigate risk factors impacting the implementation of the audit risk model and propose a complementary risk-based instrument (KRIs) to form substance risk judgments and mitigate against heightened risk of material misstatement (RMM). The study addresses the question of how risk factors impact the implementation of the audit risk model, improve risk judgments, and aid in the adoption of CAATTs. The study uses a three-stage scale development procedure involving a pretest and subsequent study with two independent samples. The pretest involves an exploratory factor analysis, while the subsequent study employs confirmatory factor analysis for construct validation. Additionally, the authors test the ability of the KRIs to predict audit efforts needed to mitigate against heightened RMM. Data was collected through two independent samples involving 767 participants. The collected data was analyzed using exploratory factor analysis and confirmatory factor analysis to assess scale validity and construct validation. The suggested KRIs, comprising two risk components and seventeen risk items, are found to have high predictive power in determining audit efforts needed to reduce RMM. The study validates the suggested KRIs as an effective instrument for risk assessment and decision-making on the adoption of CAATTs. This study contributes to the existing literature by implementing a holistic approach to risk assessment and providing a quantitative expression of assessed risks. It bridges the gap between intuitive risk evaluation and the theoretical domain, clarifying the mechanism of risk assessments. It also helps improve the uniformity and quality of risk assessments, aiding audit standard-setters in issuing updated guidelines on CAATT adoption. A few limitations and recommendations for future research should be mentioned. First, the process of developing the scale was conducted in the Israeli auditing market, which follows the International Standards on Auditing (ISAs). Although ISAs are adopted in European countries, for greater generalization, future studies could focus on other countries that adopt additional or local auditing standards. Second, this study revealed risk factors that have a material impact on the assessed risk. However, there could be additional risk factors that influence the assessment of the RMM. Therefore, future research could investigate other risk segments, such as operational and financial risks, to bring a broader generalizability to our results. Third, although the sample size in this study fits acceptable scale development procedures and enables drawing conclusions from the body of research, future research may develop standardized measures based on larger samples to reduce the generation of equivocal results and suggest an extended risk model.

Keywords: audit risk model, audit efforts, CAATTs adoption, key risk indicators, sustainability

Procedia PDF Downloads 34

Authors: Sabina Auhunas

Abstract:

These days in Ukraine, Open Science dramatically develops for the sake of scientists of all branches, providing an opportunity to take a more close look on the studies by foreign scientists, as well as to deliver their own scientific data to national and international journals. However, when it comes to the generalization of data on science activities by Ukrainian scientists, these data are often integrated into E-systems that operate inconsistent and barely related information sources. In order to resolve these issues, developed countries productively use E-systems, designed to store and manage research data, such as Current Research Information Systems that enable combining uncompiled data obtained from different sources. An algorithm for selecting SARS-CoV-2 research articles was designed, by means of which we collected the set of papers published by Ukrainian scientists and uploaded by August 1, 2020. Resulting metadata (document type, open access status, citation count, h-index, most cited documents, international research funding, author counts, the bibliographic relationship of journals) were taken from Scopus and Web of Science databases. The study also considered the info from COVID-19/SARS-CoV-2-related documents published from December 2019 to September 2020, directly from documents published by authors depending on territorial affiliation to Ukraine. These databases are enabled to get the necessary information for bibliometric analysis and necessary details: copyright, which may not be available in other databases (e.g., Science Direct). Search criteria and results for each online database were considered according to the WHO classification of the virus and the disease caused by this virus and represented (Table 1). First, we identified 89 research papers that provided us with the final data set after consolidation and removing duplication; however, only 56 papers were used for the analysis. The total number of documents by results from the WoS database came out at 21641 documents (48 affiliated to Ukraine among them) in the Scopus database came out at 32478 documents (41 affiliated to Ukraine among them). According to the publication activity of Ukrainian scientists, the following areas prevailed: Education, educational research (9 documents, 20.58%); Social Sciences, interdisciplinary (6 documents, 11.76%) and Economics (4 documents, 8.82%). The highest publication activity by institution types was reported in the Ministry of Education and Science of Ukraine (its percent of published scientific papers equals 36% or 7 documents), Danylo Halytsky Lviv National Medical University goes next (5 documents, 15%) and P. L. Shupyk National Medical Academy of Postgraduate Education (4 documents, 12%). Basically, research activities by Ukrainian scientists were funded by 5 entities: Belgian Development Cooperation, the National Institutes of Health (NIH, U.S.), The United States Department of Health & Human Services, grant from the Whitney and Betty MacMillan Center for International and Area Studies at Yale, a grant from the Yale Women Faculty Forum. Based on the results of the analysis, we obtained a set of published articles and preprints to be assessed on the variety of features in upcoming studies, including citation count, most cited documents, a bibliographic relationship of journals, reference linking. Further research on the development of the national scientific E-database continues using brand new analytical methods.

Keywords: content analysis, COVID-19, scientometrics, text mining

Procedia PDF Downloads 88

Authors: Lassi Lehto, Jaakko Kahkonen, Juha Oksanen, Tapani Sarjakoski

Abstract:

The use of raster-formatted data sets in geospatial analysis is increasing rapidly. At the same time, geographic data are being introduced into disciplines outside the traditional domain of geoinformatics, like climate change, intelligent transport, and immigration studies. These developments call for better methods to deliver raster geodata in an efficient and easy-to-use manner. Data cube technologies have traditionally been used in the geospatial domain for managing Earth Observation data sets that have strict requirements for effective handling of time series. The same approach and methodologies can also be applied in managing other types of geospatial data sets. A cloud service-based geodata cube, called GeoCubes Finland, has been developed to support online delivery and analysis of most important geospatial data sets with national coverage. The main target group of the service is the academic research institutes in the country. The most significant aspects of the GeoCubes data repository include the use of multiple resolution levels, cloud-optimized file structure, and a customized, flexible content access API. Input data sets are pre-processed while being ingested into the repository to bring them into a harmonized form in aspects like georeferencing, sampling resolutions, spatial subdivision, and value encoding. All the resolution levels are created using an appropriate generalization method, selected depending on the nature of the source data set. Multiple pre-processed resolutions enable new kinds of online analysis approaches to be introduced. Analysis processes based on interactive visual exploration can be effectively carried out, as the level of resolution most close to the visual scale can always be used. In the same way, statistical analysis can be carried out on resolution levels that best reflect the scale of the phenomenon being studied. Access times remain close to constant, independent of the scale applied in the application. The cloud service-based approach, applied in the GeoCubes Finland repository, enables analysis operations to be performed on the server platform, thus making high-performance computing facilities easily accessible. The developed GeoCubes API supports this kind of approach for online analysis. The use of cloud-optimized file structures in data storage enables the fast extraction of subareas. The access API allows for the use of vector-formatted administrative areas and user-defined polygons as definitions of subareas for data retrieval. Administrative areas of the country in four levels are available readily from the GeoCubes platform. In addition to direct delivery of raster data, the service also supports the so-called virtual file format, in which only a small text file is first downloaded. The text file contains links to the raster content on the service platform. The actual raster data is downloaded on demand, from the spatial area and resolution level required in each stage of the application. By the geodata cube approach, pre-harmonized geospatial data sets are made accessible to new categories of inexperienced users in an easy-to-use manner. At the same time, the multiresolution nature of the GeoCubes repository facilitates expert users to introduce new kinds of interactive online analysis operations.

Keywords: cloud service, geodata cube, multiresolution, raster geodata

Procedia PDF Downloads 104

Authors: Pradipta Chandra, Titas Bhattacharjee, Bhaskar Bhowmick

Abstract:

The foundation of economy in India is primarily based on agriculture while this is the most neglected in the rural setting. More significantly, household women take part in agriculture with higher involvement. However, because of lower education of women they have limited access towards financial decisions, land ownership and technology but they have vital role towards the individual family level. There are limited studies on the institution-wise training barriers with the focus of gender disparity. The main purpose of this paper is to find out the factors of institution-wise training (non-formal education) barriers in technology transfer with the focus of participation of rural women in agriculture. For this study primary and secondary data were collected in the line of qualitative and quantitative approach. Qualitative data were collected by several field visits in the adjacent areas of Seva-Bharati, Seva Bharati Krishi Vigyan Kendra through semi-structured questionnaires. In the next level detailed field surveys were conducted with close-ended questionnaires scored on the seven-point Likert scale. Sample size was considered as 162. During the data collection the focus was to include women although some biasness from the end of respondents and interviewer might exist due to dissimilarity in observation, views etc. In addition to that the heterogeneity of sample is not very high although female participation is more than fifty percent. Data were analyzed using Exploratory Factor Analysis (EFA) technique with the outcome of three significant factors of training barriers in technology adoption by farmers: (a) Failure of technology transfer training (TTT) comprehension interprets that the technology takers, i.e., farmers can’t understand the technology either language barrier or way of demonstration exhibited by the experts/ trainers. (b) Failure of TTT customization, articulates that the training for individual farmer, gender crop or season-wise is not tailored. (c) Failure of TTT generalization conveys that absence of common training methods for individual trainers for specific crops is more prominent at the community level. The central finding is that the technology transfer training method can’t fulfill the need of the farmers under an economically challenged area. The impact of such study is very high in the area of dry lateritic and resource crunch area of Jangalmahal under Paschim Medinipur district, West Bengal and areas with similar socio-economy. Towards the policy level decision this research may help in framing digital agriculture for implementation of the appropriate information technology for the farming community, effective and timely investment by the government with the selection of beneficiary, formation of farmers club/ farm science club etc. The most important research implication of this study lies upon the contribution towards the knowledge diffusion mechanism of the agricultural sector in India. Farmers may overcome the barriers to achieve higher productivity through adoption of modern farm practices. Corporates will be interested in agro-sector through investment under corporate social responsibility (CSR). The research will help in framing public or industry policy and land use pattern. Consequently, a huge mass of rural farm-women will be empowered and farmer community will be benefitted.

Keywords: dry lateritic zone, institutional barriers, technology transfer in India, farm-women participation

Procedia PDF Downloads 346

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 69