Search results for: quest based learning

Authors: Norah Alosayl

Abstract:

English is considered the most important foreign language in the Kingdom of Saudi Arabia (KSA) because of the usefulness of English as a global language compared to Arabic. As students’ desire to improve their English language skills has grown, English writing has been identified as the most difficult problem for Saudi students in their language learning. Although the English language in Saudi Arabia is taught beginning in the seventh grade, many students have problems at the university level, especially in writing, due to a gap between what is taught in secondary and high schools and university expectations- pupils generally study English at school, based on one book with few exercises in vocabulary and grammar exercises, and there are no specific writing lessons. Moreover, from personal teaching experience at King Saud bin Abdulaziz University, students face real problems with their writing. This paper revolves around the blackboard-based collaborative writing to help the undergraduate Saudi EFL students, in their first year enrolled in two sections of ENGL 101 in the first semester of 2021 at King Saud bin Abdulaziz University, practice the most difficult skill they found in their writing through a small group. Therefore, a sequential mixed methods design will be suited. The first phase of the study aims to highlight the most difficult skill experienced by students from an official writing exam that is evaluated by their teachers through an official rubric used in King Saud bin Abdulaziz University. In the second phase, this study will intend to investigate the benefits of social interaction on the process of learning writing. Students will be provided with five collaborative writing tasks via discussion feature on Blackboard to practice a skill that they found difficult in writing. the tasks will be formed based on social constructivist theory and pedagogic frameworks. The interaction will take place between peers and their teachers. The frequencies of students’ participation and the quality of their interaction will be observed through manual counting, screenshotting. This will help the researcher understand how students actively work on the task through the amount of their participation and will also distinguish the type of interaction (on task, about task, or off-task). Semi-structured interviews will be conducted with students to understand their perceptions about the blackboard-based collaborative writing tasks, and questionnaires will be distributed to identify students’ attitudes with the tasks.

Keywords: writing difficulties, blackboard-based collaborative writing, process of learning writing, interaction, participations

Procedia PDF Downloads 180

Authors: Seyed Mehdi Ghezi, Hesam Hasanpoor

Abstract:

This scientific article explores enhancement methods for ensemble learning with the aim of improving the performance of classifiers in the analysis and classification of brain signals. The research approach in this field consists of two main parts, each with its own strengths and weaknesses. The choice of approach depends on the specific research question and available resources. By combining these approaches and leveraging their respective strengths, researchers can enhance the accuracy and reliability of classification results, consequently advancing our understanding of the brain and its functions. The first approach focuses on utilizing machine learning methods to identify the best features among the vast array of features present in brain signals. The selection of features varies depending on the research objective, and different techniques have been employed for this purpose. For instance, the genetic algorithm has been used in some studies to identify the best features, while optimization methods have been utilized in others to identify the most influential features. Additionally, machine learning techniques have been applied to determine the influential electrodes in classification. Ensemble learning plays a crucial role in identifying the best features that contribute to learning, thereby improving the overall results. The second approach concentrates on designing and implementing methods for selecting the best classifier or utilizing meta-classifiers to enhance the final results in ensemble learning. In a different section of the research, a single classifier is used instead of multiple classifiers, employing different sets of features to improve the results. The article provides an in-depth examination of each technique, highlighting their advantages and limitations. By integrating these techniques, researchers can enhance the performance of classifiers in the analysis and classification of brain signals. This advancement in ensemble learning methodologies contributes to a better understanding of the brain and its functions, ultimately leading to improved accuracy and reliability in brain signal analysis and classification.

Keywords: ensemble learning, brain signals, classification, feature selection, machine learning, genetic algorithm, optimization methods, influential features, influential electrodes, meta-classifiers

Procedia PDF Downloads 66

Authors: Yasser F. Hassan

Abstract:

Social networking involves use of an online platform or website that enables people to communicate, usually for a social purpose, through a variety of services, most of which are web-based and offer opportunities for people to interact over the internet, e.g. via e-mail and ‘instant messaging’, by analyzing the voting behavior and ratings of judges in a popular comments in social networks. While most of the party literature omits the electorate, this paper presents a model where elites and parties are emergent consequences of the behavior and preferences of voters. The research in artificial intelligence and psychology has provided powerful illustrations of the way in which the emergence of intelligent behavior depends on the development of representational structure. As opposed to the classical voting system (one person – one decision – one vote) a new voting system is designed where agents with opposed preferences are endowed with a given number of votes to freely distribute them among some issues. The paper uses ideas from machine learning, artificial intelligence and soft computing to provide a model of the development of voting system response in a simulated agent. The modeled development process involves (simulated) processes of evolution, learning and representation development. The main value of the model is that it provides an illustration of how simple learning processes may lead to the formation of structure. We employ agent-based computer simulation to demonstrate the formation and interaction of coalitions that arise from individual voter preferences. We are interested in coordinating the local behavior of individual agents to provide an appropriate system-level behavior.

Keywords: voting system, rough sets, multi-agent, social networks, emergence, power indices

Procedia PDF Downloads 385

Authors: Eyibo Ebengeobong Eddie, Halil Zafer Alibaba

Abstract:

The neglect of classroom acoustics is dominant in most educational facilities, meanwhile, hearing and listening is the learning process in this kind of facilities. A classroom should therefore be an environment that encourages listening, without an obstacles to understanding what is being taught. Although different studies have shown teachers to complain that noise is the everyday factor that causes stress in classroom, the capacity of individuals to understand speech is further affected by Echoes, Reverberation, and room modes. It is therefore necessary for classrooms to have an ideal acoustics to aid the intelligibility of students in the learning process. The influence of these acoustical parameters on learning and teaching in schools needs to be further researched upon to enhance the teaching and learning capacity of both teacher and student. For this reason, there is a strong need to provide and collect data to analyse and define the suitable quality of classrooms needed for a learning environment. Research has shown that acoustical problems are still experienced in both newer and older schools. However, recently, principle of acoustics has been analysed and room acoustics can now be measured with various technologies and sound systems to improve and solve the problem of acoustics in classrooms. These acoustic solutions, materials, construction methods and integration processes would be discussed in this paper.

Keywords: classroom, acoustics, materials, integration, speech intelligibility

Procedia PDF Downloads 406

Authors: C. Zimmermann, E. Lackner, M. Ebner

Abstract:

Massive Open Online Courses (MOOCs) are increasingly popular learning hubs that are boasting considerable participant numbers, innovative technical features, and a multitude of instructional resources. Still, there is a high level of evidence showing that almost all MOOCs suffer from a declining frequency of participant activity and fairly low completion rates. In this paper, we would like to share the lessons learned in implementing several design patterns that have been suggested in order to foster participant activity. Our conclusions are based on experiences with the ‘Dr. Internet’ MOOC, which was created as an xMOOC to raise awareness for a more critical approach to online health information: participants had to diagnose medical case studies. There is a growing body of recommendations (based on Learning Analytics results from earlier xMOOCs) as to how the decline in participant activity can be alleviated. One promising focus in this regard is instructional design patterns, since they have a tremendous influence on the learner’s motivation, which in turn is a crucial trigger of learning processes. Since Medieval Age storytelling, micro-learning units and specific comprehensible, narrative structures were chosen to animate the audience to follow narration. Hence, MOOC participants are not likely to abandon a course or information channel when their curiosity is kept at a continuously high level. Critical aspects that warrant consideration in this regard include shorter course duration, a narrative structure with suspense peaks (according to the ‘storytelling’ approach), and a course schedule that is diversified and stimulating, yet easy to follow. All of these criteria have been observed within the design of the Dr. Internet MOOC: 1) the standard eight week course duration was shortened down to six weeks, 2) all six case studies had a special quiz format and a corresponding resolution video which was made available in the subsequent week, 3) two out of six case studies were split up in serial video sequences to be presented over the span of two weeks, and 4) the videos were generally scheduled in a less predictable sequence. However, the statistical results from the first run of the MOOC do not indicate any strong influences on the retention rate, so we conclude with some suggestions as to why this might be and what aspects need further consideration.

Keywords: case study, Dr. internet, experience, MOOCs, design patterns

Procedia PDF Downloads 250

Authors: Gul Rahim Rahimi

Abstract:

The IoT-based soil moisture monitoring system for indoor plants is designed to address the challenges of maintaining optimal moisture levels in soil for plant growth and health. The system utilizes sensor technology to collect real-time data on soil moisture levels, which is then processed and analyzed using machine learning algorithms. This allows for accurate and timely monitoring of soil moisture levels, ensuring plants receive the appropriate amount of water to thrive. The main objectives of the system are twofold: to keep plants fresh and healthy by preventing water deficiency and to provide users with comprehensive insights into the water content of the soil on a daily and hourly basis. By monitoring soil moisture levels, users can identify patterns and trends in water consumption, allowing for more informed decision-making regarding watering schedules and plant care. The scope of the system extends to the agriculture industry, where it can be utilized to minimize the efforts required by farmers to monitor soil moisture levels manually. By automating the process of soil moisture monitoring, farmers can optimize water usage, improve crop yields, and reduce the risk of plant diseases associated with over or under-watering. Key technologies employed in the system include the Capacitive Soil Moisture Sensor V1.2 for accurate soil moisture measurement, the Node MCU ESP8266-12E Board for data transmission and communication, and the Arduino framework for programming and development. Additionally, machine learning algorithms are utilized to analyze the collected data and provide actionable insights. Cloud storage is utilized to store and manage the data collected from multiple sensors, allowing for easy access and retrieval of information. Overall, the IoT-based soil moisture monitoring system offers a scalable and efficient solution for indoor plant care, with potential applications in agriculture and beyond. By harnessing the power of IoT and machine learning, the system empowers users to make informed decisions about plant watering, leading to healthier and more vibrant indoor environments.

Keywords: IoT-based, soil moisture monitoring, indoor plants, water management

Procedia PDF Downloads 39

Authors: Norbert Csizmadia

Abstract:

The paper is based on a thorough investigation regarding the recent global, social and geographical processes. The ‘Geofusion’ book series by the author guides the readers with the help of newly illustrated “associative” geographic maps of the global world in the 21st century through the quest for the winning nations, communities, leaders and powers of this age. Hence, the above mentioned represent the research objectives, the preliminary findings of which are presented in this paper. The most significant recognition is that scientists who are recognized as explorers, geostrategists of this century, in this case, are expected to present guidelines for our new world full of global social and economic challenges. To do so, new maps are needed which do not miss the wisdom and tools of the old but complement them with the new structure of knowledge. Using the lately discovered geographic and economic interrelations, the study behind this presentation tries to give a prognosis of the global processes. The methodology applied contains the survey and analysis of many recent publications worldwide regarding geostrategic, cultural, geographical, social, and economic surveys structured into global networks. In conclusion, the author presents the result of the study, which is a collage of the global map of the 21st century as mentioned above, and it can be considered as a potential contribution to the recent scientific literature on the topic. In summary, this paper displays the results of several-year-long research giving the audience an image of how economic navigation tools can help investors, politicians and travelers to get along in the changing new world.

Keywords: geography, economic geography, geo-fusion, geostrategy

Procedia PDF Downloads 120

Authors: Martina Holenko Dlab, Natasa Hoic-Bozic

Abstract:

This paper presents an original approach to student and group activity level assessment that relies on certainty factors theory. Activity level is used to represent quantity and continuity of student’s contributions in individual and collaborative e‑learning activities (e‑tivities) and is calculated to assist teachers in assessing quantitative aspects of student's achievements. Calculated activity levels are also used to raise awareness and provide recommendations during the learning process. The proposed approach was implemented within the educational recommender system ELARS and validated using data obtained from e‑tivity realized during a blended learning course. The results showed that the proposed approach can be used to estimate activity level in the context of e-tivities realized using Web 2.0 tools as well as to facilitate the assessment of quantitative aspect of students’ participation in e‑tivities.

Keywords: assessment, ELARS, e-learning, recommender systems, student model

Procedia PDF Downloads 252

Authors: Aanuoluwapo Olusola Adesanya, Adesina Joseph

Abstract:

This paper examined the class-size and instructional materials as correlates of pupils learning and academic achievement in primary school. The population of the study comprised 198 primary school pupils in three selected schools in Ogun State, Nigeria. Data were collected through questionnaire and were analysed with the use of multiple regression and ANOVA to analysed the correlation between class-size, instructional materials (independent variables) and learning achievement (dependent variable). The findings revealed that schools having an average class-size of 30 and below with use of instructional materials obtained better results than schools having more than 30 and above. The main score were higher in the school in schools having 30 and below than schools with 30 and above. It was therefore recommended that government, stakeholders and NGOs should provide more classrooms and supply of adequate instructional materials in all primary schools in the state to cater for small class-size.

Keywords: class-size, instructional materials, learning, academic achievement

Procedia PDF Downloads 330

Authors: Tharini N. de Silva, Xiao Zhibo, Zhao Rui, Mao Kezhi

Abstract:

Causal relation identification is a crucial task in information extraction and knowledge discovery. In this work, we present two approaches to causal relation identification. The first is a classification model trained on a set of knowledge-based features. The second is a deep learning based approach training a model using convolutional neural networks to classify causal relations. We experiment with several different convolutional neural networks (CNN) models based on previous work on relation extraction as well as our own research. Our models are able to identify both explicit and implicit causal relations as well as the direction of the causal relation. The results of our experiments show a higher accuracy than previously achieved for causal relation identification tasks.

Keywords: causal realtion extraction, relation extracton, convolutional neural network, text representation

Procedia PDF Downloads 712

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

Abstract:

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

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

Procedia PDF Downloads 126

Authors: Maria Isabel Garcia-Planas, Maria Victoria Garcia-Camba

Abstract:

The use of brain stem auditory evoked potential (BAEP) is a common way to study the auditory function of people, a way to learn the functionality of a part of the brain neuronal groups that intervene in the learning process by studying the behaviour of wave VI. The latest advances in neuroscience have revealed the existence of different brain activity in the learning process that can be highlighted through the use of innocuous, low-cost, and easy-access techniques such as, among others, the BAEP that can help us to detect early possible neurodevelopmental difficulties for their subsequent assessment and cure. To date and to the authors' best knowledge, only the latency data obtained, observing the first to V waves and mainly in the left ear, were taken into account. This work shows that it is essential to take into account both ears; with these latest data, it has been possible had diagnosed more precise some cases than with the previous data had been diagnosed as 'normal' despite showing signs of some alteration that motivated the new consultation to the specialist.

Keywords: ear, neurodevelopment, auditory evoked potentials, intervals of normality, learning disabilities

Procedia PDF Downloads 151

Authors: Stefan Papastefanou

Abstract:

Artificial intelligence (AI) is an interdisciplinary field of computer science with the aim of creating intelligent machine behavior. Early approaches to AI have been configured to operate in very constrained environments where the behavior of the AI system was previously determined by formal rules. Knowledge was presented as a set of rules that allowed the AI system to determine the results for specific problems; as a structure of if-else rules that could be traversed to find a solution to a particular problem or question. However, such rule-based systems typically have not been able to generalize beyond the knowledge provided. All over the world and especially in IT-heavy industries such as the United States, the European Union, Singapore, and China, machine learning has developed to be an immense asset, and its applications are becoming more and more significant. It has to be examined how such products of machine learning models can and should be protected by IP law and for the purpose of this paper patent law specifically, since it is the IP law regime closest to technical inventions and computing methods in technical applications. Genetic breeding models are currently less popular than recursive neural network method and deep learning, but this approach can be more easily described by referring to the evolution of natural organisms, and with increasing computational power; the genetic breeding method as a subset of the evolutionary algorithms models is expected to be regaining popularity. The research method focuses on patentability (according to the world’s most significant patent law regimes such as China, Singapore, the European Union, and the United States) of AI inventions and machine learning. Questions of the technical nature of the problem to be solved, the inventive step as such, and the question of the state of the art and the associated obviousness of the solution arise in the current patenting processes. Most importantly, and the key focus of this paper is the problem of patenting inventions that themselves are developed through machine learning. The inventor of a patent application must be a natural person or a group of persons according to the current legal situation in most patent law regimes. In order to be considered an 'inventor', a person must actually have developed part of the inventive concept. The mere application of machine learning or an AI algorithm to a particular problem should not be construed as the algorithm that contributes to a part of the inventive concept. However, when machine learning or the AI algorithm has contributed to a part of the inventive concept, there is currently a lack of clarity regarding the ownership of artificially created inventions. Since not only all European patent law regimes but also the Chinese and Singaporean patent law approaches include identical terms, this paper ultimately offers a comparative analysis of the most relevant patent law regimes.

Keywords: algorithms, inventor, genetic breeding models, machine learning, patentability

Procedia PDF Downloads 101

Authors: Mohammad Alshehri, Fahd Alfarsi

Abstract:

Efficient supply chains play an essential role in delivering humanitarian supplies and directly impact the success of public aid initiatives globally. Predicting the delivery status of these essential supplies in a timely manner is crucial. Therefore, this study explores the application of various machine learning approaches to predict whether humanitarian deliveries will be made on time, using a comprehensive case-study dataset provided by one of the largest international supplying organisations. We employed several machine learning methods such as regression-based (e.g., logistics regression), tree-based (e.g., decision trees) and ensemble-based (e.g., AdaBoost, XGBoost, and Gradient Boosting) to develop our predictive model. Our findings demonstrate that ensemble algorithms achieved promising results, with F1 scores ranging from 0.90 to 0.98. These high accuracy levels indicate the robustness of ensemble-learning techniques in forecasting delivery status, potentially enabling more proactive and efficient supply chain management in global aid initiatives. The implications of this study suggest that integrating advanced predictive analytics can significantly enhance the reliability of supply chains, ensuring the timely delivery of critical commodities to those in need.

Keywords: humanitarian aids, supply chains, machine learning, delivery status

Procedia PDF Downloads 10

Authors: Rashika Sharma

Abstract:

The campuses at tertiary institutes can act as a social environment for peer to peer connections. However, socialization is not the only aspect that campuses provide. The campus can act as a learning environment that has often been termed as the campus curriculum. Many tertiary institutes have taken steps to make their campus a ‘green campus’ whereby initiatives have been taken to reduce their impact on the environment. However, as visible as these initiatives are, it is debatable whether these have any effect on students’ and their understanding of sustainable campus operations. Therefore, research was conducted to evaluate the effectiveness of sustainable campus operations in raising students’ awareness of sustainability. Students at two vocational institutes participated in this interpretive research with data collected through surveys and focus groups. The findings indicated that majority of vocational education students remained oblivious of sustainability initiatives on campuses.

Keywords: campus learning, education for sustainability, social learning, vocational education

Procedia PDF Downloads 272

Authors: Hala Shaari, Nuredin Ahmed

Abstract:

Teaching computer programming is recognized to be difficult and a real challenge. The biggest problem faced by novice programmers is their lack of understanding of basic programming concepts. A visualized learning tool was developed and used by volunteered first-year students for two semesters. The purposes of this paper are firstly, to emphasize factors which directly affect the performance of our students negatively. Secondly, to examine whether the proposed tool would improve their performance and learning progression. The results of adopting this tool were conducted using a pre-survey and post-survey questionnaire. As a result, students who used the learning tool showed better performance in their programming subject.

Keywords: factors, novice, programming, visualization

Procedia PDF Downloads 354

Authors: Cahit Perkgöz

Abstract:

Artificial intelligence has brought new approaches to solving problems in almost every research field in recent years. One of these topics is shape design and optimization, which has the possibility of applications in many fields, such as nanotechnology and electronics. A properly constructed cost function can eliminate the need for labeled data required in deep learning and create desired shapes. In this work, the network parameters are optimized differentially, which differs from traditional approaches. The methods are tested for physics-related structures and successful results are obtained. This work is supported by Eskişehir Technical University scientific research project (Project No: 20ADP090)

Keywords: deep learning, shape design, optimization, artificial intelligence

Procedia PDF Downloads 143

Authors: Naziema Begum Jappie

Abstract:

The internationalization of higher education has become a focal point for academic institutions worldwide, including those in South Africa. This paper explores the multifaceted implications of internationalization on management and practice within the South African higher education landscape. Universities all over the world are increasingly recognizing the challenges of globalization and the pressures towards internationalization. Internationalization in higher education encompasses a range of activities, including academic exchange programs, research collaborations, joint degree programs, and the recruitment of international students and faculty. In South Africa, this process is driven by various factors, including the quest for global competitiveness, the pursuit of academic excellence, and the promotion of cultural diversity. However, while internationalization presents numerous opportunities, it also brings forth significant challenges that require careful consideration by management and practitioners in higher education institutions. Furthermore, the internationalization of higher education in South Africa has significant implications for teaching and learning practices. With an increasingly diverse student body, educators must employ innovative pedagogical approaches that cater to the needs and preferences of a multicultural cohort. This may involve the integration of global perspectives into the curriculum, the use of technology-enhanced learning platforms, and the promotion of intercultural competence among students and faculty. Additionally, the exchange of knowledge and ideas with international partners can enrich research activities and contribute to the advancement of knowledge in various fields. The internationalization of higher education in South Africa has profound implications for management and practice within academic institutions. While it offers opportunities for enhancing academic quality, promoting cultural exchange, and advancing research agendas, it also presents challenges that require strategic planning, resource allocation, and stakeholder engagement. By addressing these challenges proactively and leveraging the opportunities presented by internationalization, South African universities can position themselves as global leaders in higher education while contributing to the socio-economic development of the country and the continent at large. This paper draws together the international experience in South Africa to explore the emerging patterns of strategy and practice in internationalizing Higher Education and will highlight some critical notions of how the concepts of internationalization and globalization in the context of higher education are understood by those who lead universities and what new challenges are being created as universities seek to become more international. Institutions cannot simply have bullet points in the strategic plan for the recruitment of international students; there has to be a complete commitment to a national strategy of inclusivity. This paper will further examine the leadership styles that ensure transformation together with the goals set out for internationalization. Discussions around adding the international relations dimension to the curriculum. Addressing the issues relevant to cross-border delivery of higher education.

Keywords: challenges, higher education, internationalization, strategic focus

Procedia PDF Downloads 39

Authors: Shrabani Ghosh

Abstract:

Automatic offensive language detection in social media has become a stirring task in today's NLP. Manual Offensive language detection is tedious and laborious work where automatic methods based on machine learning are only alternatives. Previous works have done sentiment analysis over social media in different ways such as supervised, semi-supervised, and unsupervised manner. Domain adaptation in a semi-supervised way has also been explored in NLP, where the source domain and the target domain are different. In domain adaptation, the source domain usually has a large amount of labeled data, while only a limited amount of labeled data is available in the target domain. Pretrained transformers like BERT, RoBERTa models are fine-tuned to perform text classification in an unsupervised manner to perform further pre-train masked language modeling (MLM) tasks. In previous work, hate speech detection has been explored in Gab.ai, which is a free speech platform described as a platform of extremist in varying degrees in online social media. In domain adaptation process, Twitter data is used as the source domain, and Gab data is used as the target domain. The performance of domain adaptation also depends on the cross-domain similarity. Different distance measure methods such as L2 distance, cosine distance, Maximum Mean Discrepancy (MMD), Fisher Linear Discriminant (FLD), and CORAL have been used to estimate domain similarity. Certainly, in-domain distances are small, and between-domain distances are expected to be large. The previous work finding shows that pretrain masked language model (MLM) fine-tuned with a mixture of posts of source and target domain gives higher accuracy. However, in-domain performance of the hate classifier on Twitter data accuracy is 71.78%, and out-of-domain performance of the hate classifier on Gab data goes down to 56.53%. Recently self-supervised learning got a lot of attention as it is more applicable when labeled data are scarce. Few works have already been explored to apply self-supervised learning on NLP tasks such as sentiment classification. Self-supervised language representation model ALBERTA focuses on modeling inter-sentence coherence and helps downstream tasks with multi-sentence inputs. Self-supervised attention learning approach shows better performance as it exploits extracted context word in the training process. In this work, a self-supervised attention mechanism has been proposed to detect hate speech on Gab.ai. This framework initially classifies the Gab dataset in an attention-based self-supervised manner. On the next step, a semi-supervised classifier trained on the combination of labeled data from the first step and unlabeled data. The performance of the proposed framework will be compared with the results described earlier and also with optimized outcomes obtained from different optimization techniques.

Keywords: attention learning, language model, offensive language detection, self-supervised learning

Procedia PDF Downloads 97

Authors: Ting Gao, Mingyue He

Abstract:

Reconfigurable intelligent surfaces (RISs) are expected to be an important part of next-generation wireless communication networks due to their potential to reduce the hardware cost and energy consumption of millimeter Wave (mmWave) massive multiple-input multiple-output (MIMO) technology. However, owing to the lack of signal processing abilities of the RIS, the perfect channel state information (CSI) in RIS-assisted communication systems is difficult to acquire. In this paper, the uplink channel estimation for mmWave systems with a hybrid active/passive RIS architecture is studied. Specifically, a deep learning-based estimation scheme is proposed to estimate the channel between the RIS and the user. In particular, the sparse structure of the mmWave channel is exploited to formulate the channel estimation as a sparse reconstruction problem. To this end, the proposed approach is derived to obtain the distribution of non-zero entries in a sparse channel. After that, the channel is reconstructed by utilizing the least-squares (LS) algorithm and compressed sensing (CS) theory. The simulation results demonstrate that the proposed channel estimation scheme is superior to existing solutions even in low signal-to-noise ratio (SNR) environments.

Keywords: channel estimation, reconfigurable intelligent surface, wireless communication, deep learning

Procedia PDF Downloads 131

Authors: Nubwimana Rachel, Mugabowindekwe Maurice

Abstract:

Trees play a crucial role in maintaining biodiversity and providing various ecological services. Traditional methods of counting trees are time-consuming, and there is a need for more efficient techniques. However, deep learning makes it feasible to identify the multi-scale elements hidden in aerial imagery. This research focuses on the application of deep learning techniques for tree detection and counting in both forest and non-forest areas through the exploration of the deep learning application for automated tree detection and counting using satellite imagery. The objective is to identify the most effective model for automated tree counting. We used different deep learning models such as YOLOV7, SSD, and UNET, along with Generative Adversarial Networks to generate synthetic samples for training and other augmentation techniques, including Random Resized Crop, AutoAugment, and Linear Contrast Enhancement. These models were trained and fine-tuned using satellite imagery to identify and count trees. The performance of the models was assessed through multiple trials; after training and fine-tuning the models, UNET demonstrated the best performance with a validation loss of 0.1211, validation accuracy of 0.9509, and validation precision of 0.9799. This research showcases the success of deep learning in accurate tree counting through remote sensing, particularly with the UNET model. It represents a significant contribution to the field by offering an efficient and precise alternative to conventional tree-counting methods.

Keywords: remote sensing, deep learning, tree counting, image segmentation, object detection, visualization

Procedia PDF Downloads 51

Authors: Paulomi Polly Burey, Mark Lynch

Abstract:

It has been shown that laboratory activities can help cement understanding of theoretical concepts, but it is difficult to deliver such an activity to an online cohort and issues such as occupational health and safety in the students’ learning environment need to be considered. Chemistry, in particular, is one of the sciences where practical experience is beneficial for learning, however typical university experiments may not be suitable for the learning environment of a distance learner. Food provides an ideal medium for demonstrating chemical concepts, and along with a few simple physical and virtual tools provided by educators, analytical chemistry can be experienced by distance learners. Food chemistry experiments were designed to be carried out in a home-based environment that 1) Had sufficient scientific rigour and skill-building to reinforce theoretical concepts; 2) Were safe for use at home by university students and 3) Had the potential to enhance student learning by linking simple hands-on laboratory activities with high-level virtual science. Two main components of the resources were developed, a home laboratory experiment component, and a virtual laboratory component. For the home laboratory component, students were provided with laboratory kits, as well as a list of supplementary inexpensive chemical items that they could purchase from hardware stores and supermarkets. The experiments used were typical proximate analyses of food, as well as experiments focused on techniques such as spectrophotometry and chromatography. Written instructions for each experiment coupled with video laboratory demonstrations were used to train students on appropriate laboratory technique. Data that students collected in their home laboratory environment was collated across the class through shared documents, so that the group could carry out statistical analysis and experience a full laboratory experience from their own home. For the virtual laboratory component, students were able to view a laboratory safety induction and advised on good characteristics of a home laboratory space prior to carrying out their experiments. Following on from this activity, students observed laboratory demonstrations of the experimental series they would carry out in their learning environment. Finally, students were embedded in a virtual laboratory environment to experience complex chemical analyses with equipment that would be too costly and sensitive to be housed in their learning environment. To investigate the impact of the intervention, students were surveyed before and after the laboratory series to evaluate engagement and satisfaction with the course. Students were also assessed on their understanding of theoretical chemical concepts before and after the laboratory series to determine the impact on their learning. At the end of the intervention, focus groups were run to determine which aspects helped and hindered learning. It was found that the physical experiments helped students to understand laboratory technique, as well as methodology interpretation, particularly if they had not been in such a laboratory environment before. The virtual learning environment aided learning as it could be utilized for longer than a typical physical laboratory class, thus allowing further time on understanding techniques.

Keywords: chemistry, food science, future pedagogy, STEM education

Procedia PDF Downloads 158

Authors: Khaled Alotaibi

Abstract:

This paper deals with developing reasoning thinking and the intrinsic-extrinsic motivation for learning, and enhancing the academic achievement of a sample of students at Teachers' College in King Saud University. The study sample included 58 students who were divided randomly into two groups; one was an experimental group with 20 students and the other was a control group with 22 students. The following tools were used: e-courses by using I-map, Reasoning Thinking Tes, questionnaire to measure the intrinsic-extrinsic motivation for learning and an academic achievement test. Experimental group was taught using e-courses by using I-map, while the control group was taught by using traditional education. The results showed that: - There were no statistically significant differences between the experimental group and the control group in Reasoning thinking skills. - There were statistically significant differences between the experimental group and the control group in the intrinsic-extrinsic motivation for learning in favor of the experimental group. - There were statistically significant differences between the experimental group and the control group in academic achievement in favor of the experimental group.

Keywords: reasoning, thinking, intrinsic motivation, active learning

Procedia PDF Downloads 412

Authors: Mridul Sharma, Praveen Saroha

Abstract:

In today's world, everyone has become so busy in their to-do tasks and daily routine that they tend to ignore some of the basal components of our life, including exercise and diet. This comparative study analyzes the pathways of the relationship between exercise and brain plasticity and also includes another variable diet to study the effects of diet on learning by answering questions including which diet is known to be the best learning supporter and what are the recommended quantities of the same. Further, this study looks into inter-relation between diet and exercise, and also some other approach of the relation between diet and exercise on learning apart from through Brain Derived Neurotrophic Factor (BDNF).

Keywords: brain derived neurotrophic factor, brain plasticity, diet, exercise

Procedia PDF Downloads 127

Authors: Warda Khan, Remo Lachmann, Adarsh S. Garakahally

Abstract:

The increasing complexity of software and its applicability in a wide range of industries, e.g., automotive, call for enhanced quality assurance techniques. Test automation is one option to tackle the prevailing challenges by supporting test engineers with fast, parallel, and repetitive test executions. A high degree of test automation allows for a shift from mundane (manual) testing tasks to a more analytical assessment of the software under test. However, a high initial investment of test resources is required to establish test automation, which is, in most cases, a limitation to the time constraints provided for quality assurance of complex software systems. Hence, a computer-aided creation of automated test cases is crucial to increase the benefit of test automation. This paper proposes the application of machine learning for the generation of automated test cases. It is based on supervised learning to analyze test specifications and existing test implementations. The analysis facilitates the identification of patterns between test steps and their implementation with test automation components. For the test case generation, this approach exploits historical data of test automation projects. The identified patterns are the foundation to predict the implementation of unknown test case specifications. Based on this support, a test engineer solely has to review and parameterize the test automation components instead of writing them manually, resulting in a significant time reduction for establishing test automation. Compared to other generation approaches, this ML-based solution can handle different writing styles, authors, application domains, and even languages. Furthermore, test automation tools require expert knowledge by means of programming skills, whereas this approach only requires historical data to generate test cases. The proposed solution is evaluated using various multi-label evaluation criteria (EC) and two small-sized real-world systems. The most prominent EC is ‘Subset Accuracy’. The promising results show an accuracy of at least 86% for test cases, where a 1:1 relationship (Multi-Class) between test step specification and test automation component exists. For complex multi-label problems, i.e., one test step can be implemented by several components, the prediction accuracy is still at 60%. It is better than the current state-of-the-art results. It is expected the prediction quality to increase for larger systems with respective historical data. Consequently, this technique facilitates the time reduction for establishing test automation and is thereby independent of the application domain and project. As a work in progress, the next steps are to investigate incremental and active learning as additions to increase the usability of this approach, e.g., in case labelled historical data is scarce.

Keywords: machine learning, multi-class, multi-label, supervised learning, test automation

Procedia PDF Downloads 117

Authors: Mariana Paz Sajon, Paula Cecilia Primogerio, Mariana Albarracin

Abstract:

This study recovers an experience of teacher training carried out in an Undergraduate Business School from a private university in Buenos Aires, Argentina. The purpose of the project was to provide teachers with an opportunity to reflect on their teaching practices at the university. The aim of the study is to systematize lessons and challenges that emerge from this teacher training experience. A group of teachers who showed a willingness to learn teaching abilities was selected to work. They completed a formative journey working in learning communities starting from the immersion in different aspects of teaching and learning, class observations, and an individual and collaborative reflection exercise in a systematic way among colleagues. In this study, the productions of the eight teachers who are members of the learning communities are analyzed, framed in an e-portfolio that they prepared during the training journey. The analysis shows that after the process of shared reflection, traits related to powerful teaching and meaningful learning have appeared in the classes. For their part, teachers reflect having reached an awareness of their own practices, identifying strengths and opportunities for improvement, and the experience of sharing their own way and knowing the successes and failures of others was valued. It is an educational journey of pedagogical transformation of the teachers, which is infrequent in business education, which could lead to a change in teaching practices for the entire Business School. The present study involves theoretical and pedagogic aspects of education in a business school in Argentina and its flow-on implications for the workplace that may be transferred to other educational contexts.

Keywords: Argentina, learning community, meaningful learning, powerful teaching, reflective practice

Procedia PDF Downloads 204

Authors: Daban Mohammed Haji

Abstract:

This paper explores the application of constructive alignment in digital education at the University of Sulaimani, focusing specifically on the Language and Culture Center, Translation Department, and English Department. Constructive alignment, an outcome-based pedagogical framework developed by John Biggs, ensures that learning activities and assessments are directly aligned with the intended learning outcomes (ILOs). The study's findings reveal a significant gap in awareness and understanding of this pedagogical concept among lecturers. Many instructors are unfamiliar with constructive alignment, and those who have some knowledge of it face considerable challenges. These challenges include aligning learning activities and assessments with the ILOs and fostering higher-order cognitive skills as outlined in the SOLO taxonomy and revised Bloom’s taxonomy. To address this issue, the existing pedagogy center at the University of Sulaimani could play a pivotal role. This center has the potential to foster faculty development and promote the adoption of constructive alignment in online teaching. By leveraging the center's expertise and resources, a tailored program can be designed to enhance faculty understanding and application of this pedagogical framework.

Keywords: constructive alignment, student-centerdness, pedagogy, bologna process

Procedia PDF Downloads 22

Authors: Jose Gay D. Gallego

Abstract:

This study examined the learning performance in General Psychology of 297 freshmen of the CPSU-Main through the Pre and Post Tests. The instructional intervention via Universal Design for Learning (UDL) was applied to 33% (97 out of 297) of these freshmen as the Treatment Group while the 67% (200) belonged to the Control Group for traditional instructions. Statistical inferences utilized one-way Analysis of Variance for mean differences; Pearson R Correlations for bivariate relationships, and; Factor Analysis for significant components that contributed most to the Universal Design for Learning instructions. Findings showed very high levels of students’ acquired UDL skills. Results in the pre test in General Psychology, respectively, were low and average when grouped into low and high achievers. There was no significant mean difference in the acquired nine UDL components when categorized into seven colleges to generalize that between colleges they were on the same very high levels. Significant differences were found in three test areas in General Psychology in eight colleges whose students in College of teacher education taking the lead in the learning performance. Significant differences were also traced in the post test in favor of the students in the treatment group. This proved that UDL really impacted the learning performance of the low achieving students. Significant correlations were revealed between the components of UDL and General Psychology. There were twenty four significant itemized components that contributed most to UDL instructional interventions. Implications were emphasized to maximizing the principles of UDL with the contention of thoughtful planning related to the four curricular pillars of UDL: (a) instructional goals, (b) instructional delivery methods, (c) instructional materials, and (d) student assessments.

Keywords: universal design for learning, enhanced performance, teaching innovation, technology in education, social science area

Procedia PDF Downloads 261

Authors: Beatriz Lafuente Alcázar, Yash Wani, Amit J. Nimunkar

Abstract:

Cardiovascular Diseases (CVDs) are the leading cause of death globally, and around 80% of sudden cardiac deaths are due to arrhythmias or irregular heartbeats. The majority of these pathologies are revealed by either short-term or long-term alterations in the electrocardiogram (ECG) morphology. The ECG is the main diagnostic tool in cardiology. It is a non-invasive, pain free procedure that measures the heart’s electrical activity and that allows the detecting of abnormal rhythms and underlying conditions. A cardiologist can diagnose a wide range of pathologies based on ECG’s form alterations, but the human interpretation is subjective and it is contingent to error. Moreover, ECG records can be quite prolonged in time, which can further complicate visual diagnosis, and deeply retard disease detection. In this context, deep learning methods have risen as a promising strategy to extract relevant features and eliminate individual subjectivity in ECG analysis. They facilitate the computation of large sets of data and can provide early and precise diagnoses. Therefore, the cardiology field is one of the areas that can most benefit from the implementation of deep learning algorithms. In the present study, a deep learning algorithm is trained following a novel approach, using a combination of different databases as the training set. The goal of the algorithm is to achieve the detection of R-peaks in ECG signals. Its performance is further evaluated in ECG signals with different origins and features to test the model’s ability to generalize its outcomes. Performance of the model for detection of R-peaks for clean and noisy ECGs is presented. The model is able to detect R-peaks in the presence of various types of noise, and when presented with data, it has not been trained. It is expected that this approach will increase the effectiveness and capacity of cardiologists to detect divergences in the normal cardiac activity of their patients.

Keywords: arrhythmia, deep learning, electrocardiogram, machine learning, R-peaks

Procedia PDF Downloads 166

Authors: Xiaodong Li, Peng Gao, Chao-Jung Huang, Shiying Hao, Xuefeng B. Ling, Yongxia Han, Yaqi Zhang, Le Zheng, Chengyin Ye, Modi Liu, Minjie Xia, Changlin Fu, Bo Jin, Karl G. Sylvester, Eric Widen

Abstract:

Predicting the risk of Pancreatic Adenocarcinoma (PA) in advance can benefit the quality of care and potentially reduce population mortality and morbidity. The aim of this study was to develop and prospectively validate a risk prediction model to identify patients at risk of new incident PA as early as 3 months before the onset of PA in a statewide, general population in Maine. The PA prediction model was developed using Deep Neural Networks, a deep learning algorithm, with a 2-year electronic-health-record (EHR) cohort. Prospective results showed that our model identified 54.35% of all inpatient episodes of PA, and 91.20% of all PA that required subsequent chemoradiotherapy, with a lead-time of up to 3 months and a true alert of 67.62%. The risk assessment tool has attained an improved discriminative ability. It can be immediately deployed to the health system to provide automatic early warnings to adults at risk of PA. It has potential to identify personalized risk factors to facilitate customized PA interventions.

Keywords: cancer prediction, deep learning, electronic health records, pancreatic adenocarcinoma

Procedia PDF Downloads 143