Search results for: maturity classification

Authors: Tejaswi Agarwal, Amritansh Agarwal

Abstract:

Forest ecosystem plays very important role in the global carbon cycle. It stores about 80% of all above ground and 40% of all below ground terrestrial organic carbon. There is much interest in the extent of tropical forests and their rates of deforestation for two reasons: greenhouse gas contributions and the impact of profoundly negative biodiversity. Deforestation has many ecological, social and economic consequences, one of which is the loss of biological diversity. The rapid deployment of remote sensing (RS) satellites and development of RS analysis techniques in the past three decades have provided a reliable, effective, and practical way to characterize terrestrial ecosystem properties. Global estimates of tropical deforestation vary widely and range from 50,000 to 170,000km2 /yr Recent FAO tropical deforestation estimates for 1990–1995 cite 116,756km2 / yr globally. Remote Sensing can prove to be a very useful tool in monitoring of forests and associated deforestation to a sufficient level of accuracy without the need of physically surveying the forest areas as many of them are physically inaccessible. The methodology for the assessment of forest cover using digital image processing (ERDAS) has been followed. The satellite data for the study was procured from Indian institute of remote Sensing (IIRS), Dehradoon in the digital format. While procuring the satellite data, care was taken to ensure that the data was cloud free and did not belong to dry and leafless season. The Normalized Difference Vegetation Index (NDVI) has been used as a numerical indicator of the reduction in ground biomass. NDVI = (near I.R - Red)/ (near I.R + Red). After calculating the NDVI variations and associated mean, we have analysed the change in ground biomass. Through this paper, we have tried to indicate the rate of deforestation over a given period of time by comparing the forest cover at different time intervals. With the help of remote sensing and GIS techniques, it is clearly shown that the total forest cover is continuously degrading and transforming into various land use/land cover category.

Keywords: remote sensing, deforestation, supervised classification, NDVI, change detection

Procedia PDF Downloads 1203

Authors: Mark Joseph Quinto, Roan Beronilla, Guiller Damian, Eliza Camaso, Ronaldo Alberto

Abstract:

Forest inventories are essential to assess the composition, structure and distribution of forest vegetation that can be used as baseline information for management decisions. Classical forest inventory is labor intensive and time-consuming and sometimes even dangerous. The use of Light Detection and Ranging (LiDAR) in forest inventory would improve and overcome these restrictions. This study was conducted to determine the possibility of using LiDAR derived data in extracting high accuracy forest biophysical parameters and as a non-destructive method for forest status analysis of San Manual, Pangasinan. Forest resources extraction was carried out using LAS tools, GIS, Envi and .bat scripts with the available LiDAR data. The process includes the generation of derivatives such as Digital Terrain Model (DTM), Canopy Height Model (CHM) and Canopy Cover Model (CCM) in .bat scripts followed by the generation of 17 composite bands to be used in the extraction of forest classification covers using ENVI 4.8 and GIS software. The Diameter in Breast Height (DBH), Above Ground Biomass (AGB) and Carbon Stock (CS) were estimated for each classified forest cover and Tree Count Extraction was carried out using GIS. Subsequently, field validation was conducted for accuracy assessment. Results showed that the forest of San Manuel has 73% Forest Cover, which is relatively much higher as compared to the 10% canopy cover requirement. On the extracted canopy height, 80% of the tree’s height ranges from 12 m to 17 m. CS of the three forest covers based on the AGB were: 20819.59 kg/20x20 m for closed broadleaf, 8609.82 kg/20x20 m for broadleaf plantation and 15545.57 kg/20x20m for open broadleaf. Average tree counts for the tree forest plantation was 413 trees/ha. As such, the forest of San Manuel has high percent forest cover and high CS.

Keywords: carbon stock, forest inventory, LiDAR, tree count

Procedia PDF Downloads 388

Authors: Rakhyun Kim, Sungho Tae

Abstract:

Global environmental issues such as abnormal weathers due to global warming, resource depletion, and ecosystem distortions have been escalating due to rapid increase of population growth, and expansion of industrial and economic development. Accordingly, initiatives have been implemented by many countries to protect the environment through indirect regulation methods such as Environmental Product Declaration (EPD), in addition to direct regulations such as various emission standards. Following this trend, life cycle assessment (LCA) techniques that provide quantitative environmental information, such as Human Toxicity Potential (HTP), for buildings are being developed in the construction industry. However, at present, the studies on the environmental database of building materials are not sufficient to provide this support adequately. The purpose of this study is to analysis human toxicity potential of major building material production stage using life cycle assessment. For this purpose, the theoretical consideration of the life cycle assessment and environmental impact category was performed and the direction of the study was set up. That is, the major material in the global warming potential view was drawn against the building and life cycle inventory database was selected. The classification was performed about 17 kinds of substance and impact index, such as human toxicity potential, that it specifies in CML2001. The environmental impact of analysis human toxicity potential for the building material production stage was calculated through the characterization. Meanwhile, the environmental impact of building material in the same category was analyze based on the characterization impact which was calculated in this study. In this study, establishment of environmental impact coefficients of major building material by complying with ISO 14040. Through this, it is believed to effectively support the decisions of stakeholders to improve the environmental performance of buildings and provide a basis for voluntary participation of architects in environment consideration activities.

Keywords: human toxicity potential, major building material, life cycle assessment, production stage

Procedia PDF Downloads 139

Authors: Somayeh Lotfi, Manuel Eggimann, Eckhard Wagner, Radosław Mróz, Jan Deja

Abstract:

One of the main environmental challenges in the construction industry is a strong social force to decrease the bulk transport of the building materials in urban environments. Considering this fact, applying more in-situ recycling technologies for Construction and Demolition Waste (CDW) is an urgent need. The European C2CA project develops a novel concrete recycling technology that can be performed purely mechanically and in situ. The technology consists of a combination of smart demolition, gentle grinding of the crushed concrete in an autogenous mill, and a novel dry classification technology called ADR to remove the fines. The feasibility of this recycling process was examined in demonstration projects involving in total 20,000 tons of End of Life (EOL) concrete from two office towers in Groningen, The Netherlands. This paper concentrates on the second demonstration project of C2CA, where EOL concrete was recycled on an industrial site. After recycling, the properties of the produced Recycled Aggregate (RA) were investigated, and results are presented. An experimental study was carried out on mechanical and durability properties of produced Recycled Aggregate Concrete (RAC) compared to those of the Natural Aggregate Concrete (NAC). The aim was to understand the importance of RA substitution, w/c ratio and type of cement to the properties of RAC. In this regard, two series of reference concrete with strength classes of C25/30 and C45/55 were produced using natural coarse aggregates (rounded and crushed) and natural sand. The RAC series were created by replacing parts of the natural aggregate, resulting in series of concrete with 0%, 20%, 50% and 100% of RA. Results show that the concrete mix design and type of cement have a decisive effect on the properties of RAC. On the other hand, the substitution of RA even at a high percentage replacement level has a minor and manageable impact on the performance of RAC. This result is a good indication towards the feasibility of using RA in structural concrete by modifying the mix design and using a proper type of cement.

Keywords: C2CA, ADR, concrete recycling, recycled aggregate, durability

Procedia PDF Downloads 391

Authors: Amritansh Agarwal, Tejaswi Agarwal

Abstract:

Forest ecosystem plays very important role in the global carbon cycle. It stores about 80% of all above ground and 40% of all below ground terrestrial organic carbon. There is much interest in the extent of tropical forests and their rates of deforestation for two reasons: greenhouse gas contributions and the impact of profoundly negative biodiversity. Deforestation has many ecological, social and economic consequences, one of which is the loss of biological diversity. The rapid deployment of remote sensing (RS) satellites and development of RS analysis techniques in the past three decades have provided a reliable, effective, and practical way to characterize terrestrial ecosystem properties. Global estimates of tropical deforestation vary widely and range from 50,000 to 170,000 km2 /yr Recent FAO tropical deforestation estimates for 1990–1995 cite 116,756km2 / yr globally. Remote Sensing can prove to be a very useful tool in monitoring of forests and associated deforestation to a sufficient level of accuracy without the need of physically surveying the forest areas as many of them are physically inaccessible. The methodology for the assessment of forest cover using digital image processing (ERDAS) has been followed. The satellite data for the study was procured from USGS website in the digital format. While procuring the satellite data, care was taken to ensure that the data was cloud and aerosol free by making using of FLAASH atmospheric correction technique. The Normalized Difference Vegetation Index (NDVI) has been used as a numerical indicator of the reduction in ground biomass. NDVI = (near I.R - Red)/ (near I.R + Red). After calculating the NDVI variations and associated mean we have analysed the change in ground biomass. Through this paper we have tried to indicate the rate of deforestation over a given period of time by comparing the forest cover at different time intervals. With the help of remote sensing and GIS techniques it is clearly shows that the total forest cover is continuously degrading and transforming into various land use/land cover category.

Keywords: remote sensing, deforestation, supervised classification, NDVI change detection

Procedia PDF Downloads 402

Authors: Amir Shahab Shahabi, Mohsen Hasirian

Abstract:

Modern societies, marked by expansive Internet connectivity and the rise of e-commerce, are now integrated with digital platforms at an unprecedented level. The efficiency, speed, and accessibility of e-commerce have garnered a substantial consumer base. Against this backdrop, electronic banking has undergone rapid proliferation within the realm of online activities. However, this growth has inadvertently given rise to an environment conducive to illicit activities, notably electronic payment fraud, posing a formidable challenge to the domain of electronic banking. A pivotal role in upholding the integrity of electronic commerce and business transactions is played by electronic fraud detection, particularly in the context of credit cards which underscores the imperative of comprehensive research in this field. To this end, our study introduces an Attentional Bi-Stream Sequence Learner (AttBiSeL) framework that leverages attention mechanisms and recurrent networks. By incorporating bidirectional recurrent layers, specifically bidirectional Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) layers, the proposed model adeptly extracts past and future transaction sequences while accounting for the temporal flow of information in both directions. Moreover, the integration of an attention mechanism accentuates specific transactions to varying degrees, as manifested in the output of the recurrent networks. The effectiveness of the proposed approach in automatic credit card fraud classification is evaluated on the European Cardholders' Fraud Dataset. Empirical results validate that the hybrid architectural paradigm presented in this study yields enhanced accuracy compared to previous studies.

Keywords: credit card fraud, deep learning, attention mechanism, recurrent neural networks

Procedia PDF Downloads 13

Authors: Amalia Ardeljan, Lexi Frankel, Divesh Manjani, Gabriela Santizo, Maximillian Guerra, Omar Rashid

Abstract:

Introduction: Cytomegalovirus (CMV) infection is a widespread infection affecting between 60-70% of people in industrialized countries. CMV has been previously correlated with a higher incidence of Hodgkin Lymphoma compared to noninfected persons. Research regarding prior CMV infection and subsequent lymphoma development is still controversial. With limited evidence, further research is needed in order to understand the relationship between previous CMV infection and subsequent lymphoma development. This study assessed the effect of CMV infection and the incidence of lymphoma afterward. Methods: A retrospective cohort study (2010-2019) was conducted through a Health Insurance Portability and Accountability Act (HIPAA) compliant national database and conducted using International Classification of Disease (ICD) 9th,10th codes, and Current Procedural Terminology (CPT) codes. These were used to identify lymphoma diagnosis in a previously CMV infected population. Patients were matched for age range and Charlson Comorbidity Index (CCI). A chi-squared test was used to assess statistical significance. Results: A total number of 14,303 patients was obtained in the CMV infected group as well as in the control population (matched by age range and CCI score). Subsequent lymphoma development was seen at a rate of 11.44% (1,637) in the CMV group and 5.74% (822) in the control group, respectively. The difference was statistically significant by p= 2.2x10-16, odds ratio = 2.696 (95% CI 2.483- 2.927). In an attempt to stratify the population by antiviral medication exposure, the outcomes were limited by the decreased number of members exposed to antiviral medication in the control population. Conclusion: This study shows a statistically significant correlation between prior CMV infection and an increased incidence of lymphoma afterward. Further exploration is needed to identify the potential carcinogenic mechanism of CMV and whether the results are attributed to a confounding bias.

Keywords: cytomegalovirus, lymphoma, cancer, microbiology

Procedia PDF Downloads 219

Authors: Tanya Keni

Abstract:

Efforts to link culture and self, have been the focus, initially of Anthropology and later of Psychology in the first half of the 20th century. In doing so, cross-cultural researchers have endeavored to identify factors valuable for classifying cultures. One such central classification is that of individualism and collectivism which remains prominent. However, it overlooks certain other cultural dimensions that can be of interest and need attention. The current paper tries to move beyond this classic distinction, to cultures that are termed to be honor and dignity oriented. Both honor and dignity, refer to the worth of a person but bear different connotations and psychological consequences. While dignity is an independent concept of self-worth whose locus lies deep within the individual, honor is an interdependent concept that needs both personal as well as societal acknowledgment. This research takes an exploratory and qualitative approach to draw the individual, structural and contextual understanding of personal honor and dignity in broad cultures that are conceptualized as honor and dignity aimed. The aim is to understand the cultural influence on an individual’s self-worth, considering gender. 12 Focus group discussions were conducted across North India and Germany with four participants each. The research process was inspired by the approaches of social constructivism and critical realism. These discussions were transcribed and further analyzed using thematic analysis and the results have revealed differential themes for the concepts of honor and dignity. Certain dimensional similarities were also observed for both the cultural groups, however with differential usage of language. In particular, the North Indian group was seen using phrases that were oriented towards safeguarding against loss of honor or dignity. While the phrases of the German group were aligned towards worth-enhancement. The research also gives an illustration of how honor and dignity translate into behavioral practice that can exert an influence on important life decisions, especially about self and family for both males and females. In addition to these, the study also contributes to the literature on self-worth by developing the concept of ‘dignity’ for which there exists a dearth of research.

Keywords: culture, dignity, honor, self, self-worth

Procedia PDF Downloads 86

Authors: Tomas Premoli, Sareh Rowlands

Abstract:

In this study, we investigate the impact of various convolutional neural network (CNN) architectures on the accuracy of diagnosing Alzheimer’s disease (AD) using patient MRI scans. Alzheimer’s disease is a debilitating neurodegenerative disorder that affects millions worldwide. Early, accurate, and non-invasive diagnostic methods are required for providing optimal care and symptom management. Deep learning techniques, particularly CNNs, have shown great promise in enhancing this diagnostic process. We aim to contribute to the ongoing research in this field by comparing the effectiveness of different CNN architectures and providing insights for future studies. Our methodology involved preprocessing MRI data, implementing multiple CNN architectures, and evaluating the performance of each model. We employed intensity normalization, linear registration, and skull stripping for our preprocessing. The selected architectures included VGG, ResNet, and DenseNet models, all implemented using the Keras library. We employed transfer learning and trained models from scratch to compare their effectiveness. Our findings demonstrated significant differences in performance among the tested architectures, with DenseNet201 achieving the highest accuracy of 86.4%. Transfer learning proved to be helpful in improving model performance. We also identified potential areas for future research, such as experimenting with other architectures, optimizing hyperparameters, and employing fine-tuning strategies. By providing a comprehensive analysis of the selected CNN architectures, we offer a solid foundation for future research in Alzheimer’s disease diagnosis using deep learning techniques. Our study highlights the potential of CNNs as a valuable diagnostic tool and emphasizes the importance of ongoing research to develop more accurate and effective models.

Keywords: Alzheimer’s disease, convolutional neural networks, deep learning, medical imaging, MRI

Procedia PDF Downloads 73

Authors: B. L. Gadiga

Abstract:

This study assesses the major ecological zones in Nigeria with the view to understanding the spatial pattern of vegetation zones and the implications on conservation within the period of sixteen (16) years. Satellite images used for this study were acquired from the SPOT-VEGETATION between 1998 and 2013. The annual NDVI images selected for this study were derived from SPOT-4 sensor and were acquired within the same season (November) in order to reduce differences in spectral reflectance due to seasonal variations. The images were sliced into five classes based on literatures and knowledge of the area (i.e. <0.16 Non-Vegetated areas; 0.16-0.22 Sahel Savannah; 0.22-0.40 Sudan Savannah, 0.40-0.47 Guinea Savannah and >0.47 Forest Zone). Classification of the 1998 and 2013 images into forested and non forested areas showed that forested area decrease from 511,691 km² in 1998 to 478,360 km² in 2013. Differencing change detection method was performed on 1998 and 2013 NDVI images to identify areas of ecological concern. The result shows that areas undergoing vegetation degradation covers an area of 73,062 km² while areas witnessing some form restoration cover an area of 86,315 km². The result also shows that there is a weak correlation between rainfall and the vegetation zones. The non-vegetated areas have a correlation coefficient (r) of 0.0088, Sahel Savannah belt 0.1988, Sudan Savannah belt -0.3343, Guinea Savannah belt 0.0328 and Forest belt 0.2635. The low correlation can be associated with the encroachment of the Sudan Savannah belt into the forest belt of South-eastern part of the country as revealed by the image analysis. The degradation of the forest vegetation is therefore responsible for the serious erosion problems witnessed in the South-east. The study recommends constant monitoring of vegetation and strict enforcement of environmental laws in the country.

Keywords: vegetation, NDVI, SPOT-vegetation, ecology, degradation

Procedia PDF Downloads 221

Authors: M. Abdullahi, S. Sadiku, Bashar S. Mohammed, J. I. Aguwa

Abstract:

The properties of hollow sandcrete blocks produced in Minna, Nigeria are presented. Sandcrete block is made of cement, water and sand bound together in certain mix proportions. For the purpose of this work, fifty (50) commercial sandcrete block industries were visited in Minna, Nigeria to obtain block samples and aggregates used for the manufacture, and to also take inventory of the mix composition and the production process. Sieve analysis tests were conduction on the soil sample from various block industries to ascertain their quality to be used for block making. The mix ratios were also investigated. Five (5) nine inches (9’’ or 225mm) blocks were obtained from each block industry and tested for dimensional compliance and compressive strength. The result of test shows that the grading of the sand falls within the limit required by BS 882: 1990. The sand particles generally satisfy the grading requirement of overall grading and also fall in at least one of the classification of coarse grading, medium grading or fine grading. This clearly indicates that the quality of the aggregates used for the production of sandcrete blocks in Minna, Nigeria are of good quality in terms of grading and workable mix can easily be achieved to obtain high quality product. Physical examinations of the block sizes show slight deviation from the standard requirement in NIS 87:2000. Compressive strength of hollow sandcrete blocks in range of 0.12 N/mm2 to 0.54 N/mm2 was obtained which is below the recommendable value of 3.45 N/mm2 for load bearing hollow sandcrete blocks. This indicates that these blocks are below the standard for load-bearing sandcrete blocks and cannot be used as load bearing walling units. The mix composition also indicated low cement content resulting in low compressive strength. Most of the commercial block industries visited do not take curing very serious. Water were only sprinkled ones or twice before the blocks were stacked and made readily available for sale. It is recommended that a mix ratio of 1:4 to 1:6 should be used for the production of sandcrete blocks and proper curing practice should be adhered to. Blocks should also be cured for 14 days before making them available for consumers.

Keywords: compressive strength, dimensions, mix proportions, sandcrete blocks

Procedia PDF Downloads 388

Authors: M. S. Jouini, A. Al-Sumaiti, M. Tembely, K. Rahimov

Abstract:

Characterizing rock properties of carbonate reservoirs is highly challenging because of rock heterogeneities revealed at several length scales. In the last two decades, the Digital Rock Physics (DRP) approach was implemented successfully in sandstone rocks reservoirs in order to understand rock properties behaviour at the pore scale. This approach uses 3D X-ray Microtomography images to characterize pore network and also simulate rock properties from these images. Even though, DRP is able to predict realistic rock properties results in sandstone reservoirs it is still suffering from a lack of clear workflow in carbonate rocks. The main challenge is the integration of properties simulated at different scales in order to obtain the effective rock property of core plugs. In this paper, we propose several approaches to characterize absolute permeability in some carbonate core plugs samples using multi-scale numerical simulation workflow. In this study, we propose a procedure to simulate porosity and absolute permeability of a carbonate rock sample using textures of Micro-Computed Tomography images. First, we discretize X-Ray Micro-CT image into a regular grid. Then, we use a textural parametric model to classify each cell of the grid using supervised classification. The main parameters are first and second order statistics such as mean, variance, range and autocorrelations computed from sub-bands obtained after wavelet decomposition. Furthermore, we fill permeability property in each cell using two strategies based on numerical simulation values obtained locally on subsets. Finally, we simulate numerically the effective permeability using Darcy’s law simulator. Results obtained for studied carbonate sample shows good agreement with the experimental property.

Keywords: multiscale modeling, permeability, texture, micro-tomography images

Procedia PDF Downloads 183

Authors: F. Fatoye, C. E. Mbada, T. Gebrye, A. O. Ogunsola, C. Fatoye, O. Oyewole

Abstract:

Musculoskeletal disorders (MSDs) are a major cause of pain and disability. It is likely to become a greater economic and public health burden that is unnecessary. Thus, reliable prevalence figures are important for both clinicians and policy-makers to plan health care needs for those affected with the disease. This study estimated hospital based real-world prevalence of MSDs in Nigeria. A review of medical charts for adult patients attending Physiotherapy Outpatient Clinic at the Obafemi Awolowo University Teaching Hospitals Complex, Osun State, Nigeria between 2009 and 2018 was carried out to identify common MSDs including low back pain (LBP), cervical spondylosis (CSD), post immobilization stiffness (PIS), sprain, osteoarthritis (OA), and other conditions. Occupational class of the patients was determined using the International Labour Classification (ILO). Data were analysed using descriptive statistics of frequency and percentages. Overall, medical charts of 3,340 patients were reviewed within the span of ten years (2009 to 2018). Majority of the patients (62.8%) were in the middle class, and the remaining were in low class (25.1%) and high class (10.5%) category. An overall prevalence of 47.35% of MSD was found within the span of ten years. Of this, the prevalence of LBP, CSD, PIS, sprain, OA, and other conditions was 21.6%, 10%, 18.9%, 2%, 6.3%, and 41.3%, respectively. The highest (14.2%) and lowest (10.5%) prevalence of MSDs was recorded in the year of 2012 and 2018, respectively. The prevalence of MSDs is considerably high among Nigerian patients attending outpatient a physiotherapy clinic. The high prevalence of MSDs underscores the need for clinicians and decision makers to put in place appropriate strategies to reduce the prevalence of these conditions. In addition, they should plan and evaluate healthcare services to improve the health outcomes of patients with MSDs. Further studies are required to determine the economic burden of the condition and examine the clinical and cost-effectiveness of physiotherapy interventions for patients with MSDs.

Keywords: musculoskeletal disorders, Nigeria, prevalence, real world

Procedia PDF Downloads 172

Authors: Yaxin Bi, Peter Nicholl

Abstract:

The rapid advancement of Information and communication technologies (ICT) is extremely influencing every aspect of Higher Education. It has transformed traditional teaching, learning, assessment and feedback into a new era of Digital Education. This also introduces many challenges in capturing and understanding student engagement with their studies in Higher Education. The School of Computing at Ulster University has developed a Feedback And Notification (FAN) Online tool that has been used to send students links to personalized feedback on their submitted assessments and record students’ frequency of review of the shared feedback as well as the speed of collection. The feedback that the students initially receive is via a personal email directing them through to the feedback via a URL link that maps to the feedback created by the academic marker. This feedback is typically a Word or PDF report including comments and the final mark for the work submitted approximately three weeks before. When the student clicks on the link, the student’s personal feedback is viewable in the browser and they can view the contents. The FAN tool provides the academic marker with a report that includes when and how often a student viewed the feedback via the link. This paper presents an investigation into student engagement through analyzing the interaction timestamps and frequency of review by the student. We have proposed an approach to modeling interaction timestamps and use sentiment classification techniques to analyze the data collected over the last five years for a set of modules. The data studied is across a number of final years and second-year modules in the School of Computing. The paper presents the details of quantitative analysis methods and describes further their interactions with the feedback overtime on each module studied. We have projected the students into different groups of engagement based on sentiment analysis results and then provide a suggestion of early targeted intervention for the set of students seen to be under-performing via our proposed model.

Keywords: feedback, engagement, interaction modelling, sentiment analysis

Procedia PDF Downloads 103

Authors: Reza Mehrafarin, Nafiseh Mirshekari, Mahyar Mehrafarin

Abstract:

In 25 km southeast of Zabul (center of Sistan, in the east of Iran), a large hill can be seen. This hill, which is located next to the bend of the Sistan river, is known as the Tappeh Shahrestan. The length of the Tappeh Shahrestan is 1350 meters, its width is 360 meters, and its height is 20 meters, which in total reaches to 48 hectares. The capital of Sistan province was Ram Shahrestan in the Sassanid period, according to Iranian historical texts and Sassanid Pahlavi traditions. The city was abandoned because the nearby river dried up. Then another capital was built in Sistan called Zarang. But due to the long passage of time since the destruction of the city, its real location was forgotten and and some archaeologists have suggested different areas as the main location of the Ram Shahrestan. In 2018, the first archaeological field activities took place on and around the hillin order to answer this question: was Tappe Shahristan the same as Ram Shahristan, the capital of Sistan, during the Sassanid period? In order to answer this question, archaeological field activities were carried out on and around the hill. The field activities of the first season included the followings: 1- Preparation of hill topography and plan metric 3-Archaeogeophysics studies 3-Methodical study of archeology 4-Determining the range of the hill by soundings5-Documentation of the hill 6-Classification, typology, and comparison of pottery typology. The results of archaeological field activities in the first phase of Tappeh Shahrestan showed that this ancient site was the same city of Ram Shahrestan, the capital of Sistan, during the Sassanid period. The beginning of settlement in this city was the third century BC and the time of leaving was the end of the third century AD. The most important factors in the creation of the city was the abundant water of the Sistan River and its convenient location, and the most important reason for the abandonment of the city was the Sistan River, whose water completely dried up.

Keywords: archaeological surveys, archaeological soundings, ram shahrestan, sistan, tappeh shahrestan

Procedia PDF Downloads 110

Authors: Induja R. Nimma, Anand Reddy Maligireddy, Artur Schneider, Melissa Lyle

Abstract:

Background: Although heart failure is one of the most common causes of hospitalization in adult patients, there is limited knowledge on outcomes following initial hospitalization for COVID-19 with heart failure (HCF-19). We felt it pertinent to analyze 30-day readmission causes and outcomes among patients with HCF-19 using the United States using real-world big data via the National readmission database. Objective: The aim is to describe the rate and causes of readmissions and morbidity of heart failure with coinciding COVID-19 (HFC-19) in the United States, using the 2020 National Readmission Database (NRD). Methods: A descriptive, retrospective study was conducted on the 2020 NRD, a nationally representative sample of all US hospitalizations. Adult (>18 years) inpatient admissions with COVID-19 with HF and readmissions in 30 days were selected based on the International Classification of Diseases-Tenth Revision, Procedure Code. Results: In 2020, 2,60,372 adult patients were hospitalized with COVID-19 and HF. The median age was 74 (IQR: 64-83), and 47% were female. The median length of stay was 7(4-13) days, and the total cost of stay was 62,025 (31,956 – 130,670) United States dollars, respectively. Among the index hospital admissions, 61,527 (23.6%) died, and 22,794 (11.5%) were readmitted within 30 days. The median age of patients readmitted in 30 days was 73 (63-82), 45% were female, and 1,962 (16%) died. The most common principal diagnosis for readmission in these patients was COVID-19= 34.8%, Sepsis= 16.5%, HF = 7.1%, AKI = 2.2%, respiratory failure with hypoxia =1.7%, and Pneumonia = 1%. Conclusion: The rate of readmission in patients with heart failure exacerbations is increasing yearly. COVID-19 was observed to be the most common principal diagnosis in patients readmitted within 30 days. Complicated hypertension, chronic pulmonary disease, complicated diabetes, renal failure, alcohol use, drug use, and peripheral vascular disorders are risk factors associated with readmission. Familiarity with the most common causes and predictors for readmission helps guide the development of initiatives to minimize adverse outcomes and the cost of medical care.

Keywords: Covid-19, heart failure, national readmission database, readmission outcomes

Procedia PDF Downloads 79

Authors: Mathew Wakefield, Matthew Mitchell, Lisa Wise, Christopher McCarthy

Abstract:

The properties of memory representations in artificial neural networks have cognitive implications. Distributed representations that encode instances as a pattern of activity across layers of nodes afford memory compression and enforce the selection of a single point in instance space. These encoding schemes also appear to distort the representational space, as well as trading off the ability to validate that input information is within the bounds of past experience. In contrast, a localist representation which encodes some meaningful information into individual nodes in a network layer affords less memory compression while retaining the integrity of the representational space. This allows the validity of an input to be determined. The validity (or familiarity) of input along with the capacity of localist representation for multiple instance selections affords a memory sampling approach that dynamically balances the bias-variance trade-off. When the input is familiar, bias may be high by referring only to the most similar instances in memory. When the input is less familiar, variance can be increased by referring to more instances that capture a broader range of features. Using this approach in a localist instance memory network, an experiment demonstrates a relationship between representational conflict, generalization performance, and memorization demand. Relatively small sampling ranges produce the best performance on a classic machine learning dataset of visual objects. Combining memory validity with conflict detection produces a reliable confidence judgement that can separate responses with high and low error rates. Confidence can also be used to signal the need for supervisory input. Using this judgement, the need for supervised learning as well as memory encoding can be substantially reduced with only a trivial detriment to classification performance.

Keywords: artificial neural networks, representation, memory, conflict monitoring, confidence

Procedia PDF Downloads 127

Authors: Dalya M. Hassan

Abstract:

Marketplaces and commercial streets represent some of the most used and lively urban public spaces. Not only do they provide an outlet for commercial exchange, but they also facilitate social and recreational encounters. Such encounters can be influenced by both formal as well as informal vending activities. This paper explores and documents forms of informal vending practices and how they relate to social patterns that occur along the sidewalks of Commercial Streets in Cairo. A qualitative single case study approach of ‘Midan El Gami’ marketplace in Heliopolis, Cairo is adopted. The methodology applied includes direct and walk-by observations for two main commercial streets in the marketplace. Four zoomed-in activity maps are also done for three sidewalk segments that displayed varying vending and social features. Main findings include a documentation and classification of types of informal vending practices as well as a documentation of vendors’ distribution patterns in the urban space. Informal vending activities mainly included informal street vendors and shop spillovers, either as product or seating spillovers. Results indicated that staying and lingering activities were more prevalent in sidewalks that had certain physical features, such as diversity of shops, shaded areas, open frontages, and product or seating spillovers. Moreover, differences in social activity patterns were noted between sidewalks with street vendors and sidewalks with spillovers. While the first displayed more buying, selling, and people watching activities, the latter displayed more social relations and bonds amongst traders’ communities and café patrons. Ultimately, this paper provides a documentation, which suggests that informal vending can have a positive influence on creating a lively commercial street and on resulting patterns of use on the sidewalk space. The results can provide a basis for further investigations and analysis concerning this topic. This could aid in better accommodating informal vending activities within the design of future commercial streets.

Keywords: commercial streets, informal vending practices, sidewalks, social encounters

Procedia PDF Downloads 163

Authors: Zachary Pan

Abstract:

Our general mission is to solve the problem of classifying images into different body part types and deciding if each of them is healthy or not. However, for now, we will determine healthiness for only one-sixth of the body parts, specifically the chest. We will detect pneumonia in X-ray scans of those chest images. With this type of AI, doctors can use it as a second opinion when they are taking CT or X-ray scans of their patients. Another ad-vantage of using this machine learning classifier is that it has no human weaknesses like fatigue. The overall ap-proach to this problem is to split the problem into two parts: first, classify the image, then determine if it is healthy. In order to classify the image into a specific body part class, the body parts dataset must be split into test and training sets. We can then use many models, like neural networks or logistic regression models, and fit them using the training set. Now, using the test set, we can obtain a realistic accuracy the models will have on images in the real world since these testing images have never been seen by the models before. In order to increase this testing accuracy, we can also apply many complex algorithms to the models, like multiplicative weight update. For the second part of the problem, to determine if the body part is healthy, we can have another dataset consisting of healthy and non-healthy images of the specific body part and once again split that into the test and training sets. We then use another neural network to train on those training set images and use the testing set to figure out its accuracy. We will do this process only for the chest images. A major conclusion reached is that convolutional neural networks are the most reliable and accurate at image classification. In classifying the images, the logistic regression model, the neural network, neural networks with multiplicative weight update, neural networks with the black box algorithm, and the convolutional neural network achieved 96.83 percent accuracy, 97.33 percent accuracy, 97.83 percent accuracy, 96.67 percent accuracy, and 98.83 percent accuracy, respectively. On the other hand, the overall accuracy of the model that de-termines if the images are healthy or not is around 78.37 percent accuracy.

Keywords: body part, healthcare, machine learning, neural networks

Procedia PDF Downloads 103

Authors: Jang-Hwan Jo

Abstract:

The supply management of ecosystem services of local forests is an essential issue as it is linked to the ecological welfare of local residents. This study aims to estimate the supply, trade-offs, and synergies of local forest regulating ecosystem services using a land cover classification map (LCCM) and a forest types map (FTM). Rigorous literature reviews and Expert Delphi analysis were conducted using the detailed variables of 1:5,000 LCCM and FTM. Land-use scoring method and Getis-Ord Gi* Analysis were utilized on detailed variables to propose a method for estimating supply, trade-offs, and synergies of the local forest regulating ecosystem services. The analysis revealed that the rank order (1st to 5th) of supply of regulating ecosystem services was Erosion prevention, Air quality regulation, Heat island mitigation, Water quality regulation, and Carbon storage. When analyzing the correlation between defined services of the entire city, almost all services showed a synergistic effect. However, when analyzing locally, trade-off effects (Heat island mitigation – Air quality regulation, Water quality regulation – Air quality regulation) appeared in the eastern and northwestern forest areas. This suggests the need to consider not only the synergy and trade-offs of the entire forest between specific ecosystem services but also the synergy and trade-offs of local areas in managing the regulating ecosystem services of local forests. The study result can provide primary data for the stakeholders to determine the initial conditions of the planning stage when discussing the establishment of policies related to the adjustment of the supply of regulating ecosystem services of the forests with limited access. Moreover, the study result can also help refine the estimation of the supply of the regulating ecosystem services with the availability of other forms of data.

Keywords: ecosystem service, getis ord gi* analysis, land use scoring method, regional forest, regulating service, synergies, trade-offs

Procedia PDF Downloads 87

Authors: Mikhail Panes, Sergei Pozniak, Nikolai Pozniak

Abstract:

Purpose: To evaluate the effectiveness of intrastromal donor limbal segments implantation for treatment of progressive keratoconus considering on main characteristics of corneal endothelial cells. Setting: Outpatient ophthalmic clinic. Methods: Twenty patients (20 eyes) with progressive keratoconus II-III of Amsler classification were recruited. The worst eye was treated with the transplantation of donor limbal segments in the recipient corneal stroma, while the fellow eye was left untreated as a control of functional and morphological changes. Furthermore, twenty patients (20 eyes) without progressive keratoconus was used as a control of corneal endothelial cells changes. All patients underwent a complete ocular examination including uncorrected and corrected distance visual acuity (UDVA, CDVA), slit lamp examination fundus examination, corneal topography and pachymetry, auto-keratometry, Anterior Segment Optical Coherence Tomography and Corneal Endothelial Specular Microscopy. Results: After two years, statistically significant improvement in the UDVA and CDVA (on the average on two lines for UDVA and three-four lines for CDVA) were noted. Besides corneal astigmatism decreased from 5.82 ± 2.64 to 1.92 ± 1.4 D. Moreover there were no statistically significant differences in the changes of mean spherical equivalent, keratometry and pachymetry indicators. It should be noted that after two years there were no significant differences in the changes of the number and form of corneal endothelial cells. It can be regarded as a process stabilization. In untreated control eyes, there was a general trend towards worsening of UDVA, CDVA and corneal thickness, while corneal astigmatism was increased. Conclusion: Intrastromal donor segments implantation is a safe technique for keratoconus treatment. Intrastromal donor segments implantation is an efficient procedure to stabilize and improve progressive keratoconus.

Keywords: corneal endothelial cells, intrastromal donor limbal segments, progressive keratoconus, surgical treatment of keratoconus

Procedia PDF Downloads 281

Authors: Francesca di Donato, Sara Trucco

Abstract:

Green accounting is a growing topic of interest. Indeed, nowadays, most firms affect the environment; therefore, companies are seeking the best way to disclose environmental information. Furthermore, companies are increasingly committed to improving the environment, and the topic is gaining more importance to the public, governments, and policymakers. Green accounting is a type of accounting that considers environmental costs and their impact on the financial performance of firms. Thus, the motivation of the current research is to investigate the state-of-the-art literature on the relationship between green accounting and firm performance since the birth of the topic of green accounting and to investigate gaps in the literature that represent fruitful terrain for future research. In doing so, this study provides a bibliometric literature review of existing evidence related to the link between green accounting and firm performance since 2000. The search, based on the most relevant databases for scientific journals (which are Scopus, Emerald, Web of Science, Google Scholar, and Econlit), returned 1917 scientific articles. The articles were manually reviewed in order to identify only the relevant studies in the field by excluding articles with titles and abstracts out of scope. The final sample was composed of 107 articles. A content analysis was carried out on the final sample of articles; in doing so, a classification system has been proposed. Findings show the most relevant environmental costs and issues considered in previous studies and how green accounting may be linked to the financial and non-financial performance of a firm. The study also offers suggestions for future research in this domain. This study has several practical implications. Indeed, the topic of green accounting may be applied to different sectors and different types of companies. Therefore, this study may help managers to better understand the most relevant environmental information to disclose and how environmental issues may be managed to improve the performance of the firms. Moreover, the bibliometric literature review may be of interest to those stakeholders who are interested in the historical evolution of the topic.

Keywords: bibliometric literature review, firm performance, green accounting, literature review

Procedia PDF Downloads 69

Authors: Laura Sour

Abstract:

Government accounting is an essential instrument of transparency and accountability in public administration, which allows connecting internal management with the implementation of policies and their evaluation by third parties through the construction of indicators on the cost of government. Several countries have adopted the International Public Sector Accounting Standards (IPSAS) as part of their modernization strategy. This document will evaluate the quantity and harmonization of the financial information published in the financial statements of 12 Latin American countries based on what is established in IPSAS 1, 2 and 17. For this, seven types of financial statements are analyzed. published during the period from 2015 to 2019. Based on this information, it will be possible to describe the evolution in the government financial publication to carry out a detailed analysis of the items that have been most transparent in these countries. Finally, the level of harmonization of the financial statements will be studied using the Herfindahl-Hirschman index (IHH) to determine the degree of comparability of the information. To date, the results indicate that the public sector has increased the quantity and harmonization of the financial information published during the study period, but in a heterogeneous way: From the data collected, it has been found that the financial statement published with greater frequency and quantity is the Income Statement (classification of expenses by nature). On the other hand, the most complete reports were published by Costa Rica (2017 to 2019) and Mexico (2016 to 2018), periods during which these countries complied with 92.9 percent of the items analyzed. Although 2017 and 2018 are the years in which the most financial statements were reported, it is important to mention that Mexico is the country that has published the most financial information throughout the entire study period. The use of the IHH is expected to provide accurate information on the quality with which countries have adopted IPSAS within their government accounting systems to promote transparency and accountability in the continent.

Keywords: accounting and auditing, government policy and regulation, harmonization, public sector accounting and audits IPSAS

Procedia PDF Downloads 90

Authors: Mirza Waseem Abbas, Syed Danish Raza

Abstract:

For many countries agriculture is still the major force driving the economy and a critically important socioeconomic sector, despite exceptional industrial development across the globe. In countries like Pakistan, this sector is considered the backbone of the economy, and most of the economic decision making revolves around agricultural outputs and data. Timely and accurate facts and figures about this vital sector hold immense significance and have serious implications for the long-term development of the economy. Therefore, any significant improvements in the statistics and other forms of data regarding agriculture sector are considered important by all policymakers. This is especially true for decision making for the betterment of crops and the agriculture sector in general. Provincial and federal agricultural departments collect data for all cash and non-cash crops and the sector, in general, every year. Traditional data collection for such a large sector i.e. agriculture, being time-consuming, prone to human error and labor-intensive, is slowly but gradually being replaced by remote sensing techniques. For this study, remotely sensed data were used for change detection (machine learning, supervised & unsupervised classification) to assess the increase or decrease in area under agriculture over the last fifteen years due to urbanization. Detailed Landsat Images for the selected agricultural districts were acquired for the year 2000 and compared to images of the same area acquired for the year 2016. Observed differences validated through detailed analysis of the areas show that there was a considerable decrease in vegetation during the last fifteen years in four major agricultural districts of the Punjab province due to urbanization (housing societies).

Keywords: change detection, area estimation, machine learning, urbanization, remote sensing

Procedia PDF Downloads 249

Authors: Kuo-Chun Chien, Chia-Hui Chang, Ren-Der Sun

Abstract:

Legal Judgment Prediction (LJP) is a subtask for legal AI. Its main purpose is to use the facts of a case to predict the judgment result. In Taiwan's criminal procedure, when prosecutors complete the investigation of the case, they will decide whether to prosecute the suspect and which article of criminal law should be used based on the facts and evidence of the case. In this study, we collected 305,240 indictments from the public inquiry system of the procuratorate of the Ministry of Justice, which included 169 charges and 317 articles from 21 laws. We take the crime facts in the indictments as the main input to jointly learn the prediction model for law source, article, and charge simultaneously based on the pre-trained Bert model. For single article cases where the frequency of the charge and article are greater than 50, the prediction performance of law sources, articles, and charges reach 97.66, 92.22, and 60.52 macro-f1, respectively. To understand the big performance gap between articles and charges, we used a bipartite graph to visualize the relationship between the articles and charges, and found that the reason for the poor prediction performance was actually due to the wording precision. Some charges use the simplest words, while others may include the perpetrator or the result to make the charges more specific. For example, Article 284 of the Criminal Law may be indicted as “negligent injury”, "negligent death”, "business injury", "driving business injury", or "non-driving business injury". As another example, Article 10 of the Drug Hazard Control Regulations can be charged as “Drug Control Regulations” or “Drug Hazard Control Regulations”. In order to solve the above problems and more accurately predict the article and charge, we plan to include the article content or charge names in the input, and use the sentence-pair classification method for question-answer problems in the BERT model to improve the performance. We will also consider a sequence-to-sequence approach to charge prediction.

Keywords: legal judgment prediction, deep learning, natural language processing, BERT, data visualization

Procedia PDF Downloads 121

Authors: Sara Alhabli, Eman Elashiry, Osama Felemban, Abdullah Almushayt, Faisal Dardeer, Ahmed Mohammad, Fajr Orri, Nada Bamashmous

Abstract:

Background: In December of 2019, the coronavirus (SARS-CoV-2) first appeared and quickly spread to become a worldwide pandemic. This study aimed to evaluate the prevalence and types of pediatric dental emergencies during the COVID-19 lockdown in Jeddah, Saudi Arabia, at the University Dental Hospital (UDH) of King Abdulaziz University (KAU) and identified the management provided for these dental emergency visits. Materials and Methods: Data collection was done retrospectively from electronic dental records for children aged 0-18 that attended the UDH emergency clinic during the period from March 1st, 2020, to September 30th, 2020. An electronic form formulated specifically for this study was used to collect the required data from electronic patient records, including demographic data, emergency classification, management, and referrals. Results: A total of 3146 patients were seen at the emergency clinics during this period, of which 661 were children (21%). Types of emergency conditions included 0.8% emergency cases, 34% urgent, and 65.2% non-urgent conditions. Severe dental pain (73.1%) and abscesses (20%) were the most common urgent dental conditions. Most non-urgent conditions presented for initial or periodic visits, recalls, or routine radiographs (74%). Treatments rarely involved restorations, with 8% among urgent conditions and 5.4% among non-urgent conditions. Antibiotics were only prescribed to 6.9% of urgent conditions. Conclusions: The largest group of children presenting at the emergency dental clinics were found to be children with non-urgent conditions. Tele dentistry can be a solution to avoid large numbers of non-urgent patients presenting to emergency clinics. Additionally, dental care for non-urgent conditions during the pandemic should focus more on procedures with less aerosol generation.

Keywords: COVID-19 pandemic, dental emergencies, oral health, pediatric dentistry, children

Procedia PDF Downloads 97

Authors: Riccardo Patriarca, Giulio Di Gravio, Francesco Costantino, Massimo Tronci

Abstract:

An accurate inventory management policy acquires a crucial role in the several high-availability sectors. In these sectors, due to the high-cost of spares and backorders, an (S-1, S) replenishment policy is necessary for high-availability items. The policy enables the shipment of a substitute efficient item anytime the inventory size decreases by one. This policy can be modelled following the Multi-Echelon Technique for Recoverable Item Control (METRIC). The METRIC is a system-based technique that allows defining the optimum stock level in a multi-echelon network, adopting measures in line with the decision-maker’s perspective. The METRIC defines an availability-cost function with inventory costs and required service levels, using as inputs data about the demand trend, the supplying and maintenance characteristics of the network and the budget/availability constraints. The traditional METRIC relies on the hypothesis that a Poisson distribution well represents the demand distribution in case of items with a low failure rate. However, in this research, we will explore the effects of using a Poisson distribution to model the demand of low failure rate items characterized by an irregular demand trend. This characteristic of a demand is not included in the traditional METRIC formulation leading to the need of revising its traditional formulation. Using the CV (Coefficient of Variation) and ADI (Average inter-Demand Interval) classification, we will define the inherent flaws of Poisson-based METRIC for irregular demand items, defining an innovative ad hoc distribution which can better fit the irregular demands. This distribution will allow defining proper stock levels to reduce stocking and backorder costs due to the high irregularities in the demand trend. A case study in the aviation domain will clarify the benefits of this innovative METRIC approach.

Keywords: METRIC, inventory management, irregular demand, spare parts

Procedia PDF Downloads 347

Authors: Manish Pandey, Gurinderjit Kaur, Meenu Talwar, Sachin Chauhan, Jagbir Gill

Abstract:

Health-care management systems are a unit of nice connection as a result of the supply a straightforward and fast management of all aspects relating to a patient, not essentially medical. What is more, there are unit additional and additional cases of pathologies during which diagnosing and treatment may be solely allotted by victimization medical imaging techniques. With associate ever-increasing prevalence, medical pictures area unit directly acquired in or regenerate into digital type, for his or her storage additionally as sequent retrieval and process. Data Mining is the process of extracting information from large data sets through using algorithms and Techniques drawn from the field of Statistics, Machine Learning and Data Base Management Systems. Forecasting may be a prediction of what's going to occur within the future, associated it's an unsure method. Owing to the uncertainty, the accuracy of a forecast is as vital because the outcome foretold by foretelling the freelance variables. A forecast management should be wont to establish if the accuracy of the forecast is within satisfactory limits. Fuzzy regression strategies have normally been wont to develop shopper preferences models that correlate the engineering characteristics with shopper preferences relating to a replacement product; the patron preference models offer a platform, wherever by product developers will decide the engineering characteristics so as to satisfy shopper preferences before developing the merchandise. Recent analysis shows that these fuzzy regression strategies area units normally will not to model client preferences. We tend to propose a Testing the strength of Exponential Regression Model over regression toward the mean Model.

Keywords: health-care management systems, fuzzy regression, data mining, forecasting, fuzzy membership function

Procedia PDF Downloads 279

Authors: Maad Kamal Al-Anni

Abstract:

This study is about an algorithmic dependence of Artificial Neural Network on Multilayer Perceptron (MPL) pertaining to the classification and clustering presentations for Mobile Adhoc Network vulnerabilities. Moreover, mobile ad hoc network (MANET) is ubiquitous intelligent internetworking devices in which it has the ability to detect their environment using an autonomous system of mobile nodes that are connected via wireless links. Security affairs are the most important subject in MANET due to the easy penetrative scenarios occurred in such an auto configuration network. One of the powerful techniques used for inspecting the network packets is Intrusion Detection System (IDS); in this article, we are going to show the effectiveness of artificial neural networks used as a machine learning along with stochastic approach (information gain) to classify the malicious behaviors in simulated network with respect to different IDS techniques. The monitoring agent is responsible for detection inference engine, the audit data is collected from collecting agent by simulating the node attack and contrasted outputs with normal behaviors of the framework, whenever. In the event that there is any deviation from the ordinary behaviors then the monitoring agent is considered this event as an attack , in this article we are going to demonstrate the  signature-based IDS approach in a MANET by implementing the back propagation algorithm over ensemble-based Traffic Table (TT), thus the signature of malicious behaviors or undesirable activities are often significantly prognosticated and efficiently figured out, by increasing the parametric set-up of Back propagation algorithm during the experimental results which empirically shown its effectiveness  for the ratio of detection index up to 98.6 percentage. Consequently it is proved in empirical results in this article, the performance matrices are also being included in this article with Xgraph screen show by different through puts like Packet Delivery Ratio (PDR), Through Put(TP), and Average Delay(AD).

Keywords: Intrusion Detection System (IDS), Mobile Adhoc Networks (MANET), Back Propagation Algorithm (BPA), Neural Networks (NN)

Procedia PDF Downloads 194

Authors: Wade Ghribi, Abdelmoty M. Ahmed, Ahmed Said Badawy, Belgacem Bouallegue

Abstract:

In higher education institutions, the most pressing priority is to improve student performance and retention. Large volumes of student data are used in Educational Data Mining techniques to find new hidden information from students' learning behavior, particularly to uncover the early symptom of at-risk pupils. On the other hand, data with noise, outliers, and irrelevant information may provide incorrect conclusions. By identifying features of students' data that have the potential to improve performance prediction results, comparing and identifying the most appropriate ensemble learning technique after preprocessing the data, and optimizing the hyperparameters, this paper aims to develop a reliable students' performance prediction model for Higher Education Institutions. Data was gathered from two different systems: a student information system and an e-learning system for undergraduate students in the College of Computer Science of a Saudi Arabian State University. The cases of 4413 students were used in this article. The process includes data collection, data integration, data preprocessing (such as cleaning, normalization, and transformation), feature selection, pattern extraction, and, finally, model optimization and assessment. Random Forest, Bagging, Stacking, Majority Vote, and two types of Boosting techniques, AdaBoost and XGBoost, are ensemble learning approaches, whereas Decision Tree, Support Vector Machine, and Artificial Neural Network are supervised learning techniques. Hyperparameters for ensemble learning systems will be fine-tuned to provide enhanced performance and optimal output. The findings imply that combining features of students' behavior from e-learning and students' information systems using Majority Vote produced better outcomes than the other ensemble techniques.

Keywords: educational data mining, student performance prediction, e-learning, classification, ensemble learning, higher education

Procedia PDF Downloads 107