Search results for: imbalanced datasets

Authors: Pius Marthin, Nihal Ata Tutkun

Abstract:

Survival analysis has become one of the paramount procedures in the modeling of time-to-event data. When we encounter complex survival problems, the traditional approach remains limited in accounting for the complex correlational structure between the covariates and the outcome due to the strong assumptions that limit the inference and prediction ability of the resulting models. Several studies exist on the deep learning approach to survival modeling; moreover, the application for the case of complex survival problems still needs to be improved. In addition, the existing models need to address the data structure's complexity fully and are subject to noise and redundant information. In this study, we design a deep learning technique (CmpXRnnSurv_AE) that obliterates the limitations imposed by traditional approaches and addresses the above issues to jointly predict the risk-specific probabilities and survival function for recurrent events with competing risks. We introduce the component termed Risks Information Weights (RIW) as an attention mechanism to compute the weighted cumulative incidence function (WCIF) and an external auto-encoder (ExternalAE) as a feature selector to extract complex characteristics among the set of covariates responsible for the cause-specific events. We train our model using synthetic and real data sets and employ the appropriate metrics for complex survival models for evaluation. As benchmarks, we selected both traditional and machine learning models and our model demonstrates better performance across all datasets.

Keywords: cumulative incidence function (CIF), risk information weight (RIW), autoencoders (AE), survival analysis, recurrent events with competing risks, recurrent neural networks (RNN), long short-term memory (LSTM), self-attention, multilayers perceptrons (MLPs)

Procedia PDF Downloads 91

Authors: Davies Obaromi, Qin Yongsong, James Ndege

Abstract:

To interpolate scattered or regularly distributed data, there are imprecise or exact methods. However, there are some of these methods that could be used for interpolating data in a regular grid and others in an irregular grid. In spatial epidemiology, it is important to examine how a disease prevalence rates are distributed in space, and how they relate with each other within a defined distance and direction. In this study, for the geographic and graphic representation of the disease prevalence, linear and biharmonic spline methods were implemented in MATLAB, and used to identify, localize and compare for smoothing in the distribution patterns of tuberculosis (TB) in Eastern Cape Province. The aim of this study is to produce a more “smooth” graphical disease map for TB prevalence patterns by a 3-D curve fitting techniques, especially the biharmonic splines that can suppress noise easily, by seeking a least-squares fit rather than exact interpolation. The datasets are represented generally as a 3D or XYZ triplets, where X and Y are the spatial coordinates and Z is the variable of interest and in this case, TB counts in the province. This smoothing spline is a method of fitting a smooth curve to a set of noisy observations using a spline function, and it has also become the conventional method for its high precision, simplicity and flexibility. Surface and contour plots are produced for the TB prevalence at the provincial level for 2012 – 2015. From the results, the general outlook of all the fittings showed a systematic pattern in the distribution of TB cases in the province and this is consistent with some spatial statistical analyses carried out in the province. This new method is rarely used in disease mapping applications, but it has a superior advantage to be assessed at subjective locations rather than only on a rectangular grid as seen in most traditional GIS methods of geospatial analyses.

Keywords: linear, biharmonic splines, tuberculosis, South Africa

Procedia PDF Downloads 240

Authors: Ali Al-Ghalib, Fouad Mohammad

Abstract:

The identification of damage in reinforced concrete structures subjected to incremental cracking performance exploiting vibration data is recognized as a challenging topic in the published and heavily cited literature. Therefore, this paper attempts to shine light on the extent of dynamic methods when applied to reinforced concrete beams simulated with various scenarios of defects. For this purpose, three different reinforced concrete beams are tested through the course of the study. The three beams are loaded statically to failure in incremental successive load cycles and later rehabilitated. After each static load stage, the beams are tested under free-free support condition using experimental modal analysis. The beams were all of the same length and cross-sectional area (2.0x0.14x0.09)m, but they were different in concrete compressive strength and the type of damage presented. The experimental modal parameters as damage identification parameters were showed computationally expensive, time consuming and require substantial inputs and considerable expertise. Nonetheless, they were proved plausible for the condition monitoring of the current case study as well as structural changes in the course of progressive loads. It was accentuated that a satisfactory localization and quantification for structural changes (Level 2 and Level 3 of damage identification problem) can only be achieved reasonably through considering frequencies and mode shapes of a system in a proper analytical model. A convenient post analysis process for various datasets of vibration measurements for the three beams is conducted in order to extract, check and correlate the basic modal parameters; namely, natural frequency, modal damping and mode shapes. The results of the extracted modal parameters and their combination are utilized and discussed in this research as quantification parameters.

Keywords: experimental modal analysis, damage identification, structural health monitoring, reinforced concrete beam

Procedia PDF Downloads 264

Authors: Peace Liz Sasha Musonge

Abstract:

Behavioral aspects of species nutrition such as feeding methods and food type are archetypal biological traits signifying how species have adapted to their environment. This concept of functional feeding groups (FFG) analysis is currently used to ascertain the trophic levels of the aquatic food web in a specific microhabitat. However, in Eastern Africa, information about the FFG classification of benthic macroinvertebrates in highland rivers and streams is almost absent, and existing studies have fragmented datasets. For this reason, we carried out a robust study to determine the feed type, trophic level and FFGs, of 56 macroinvertebrate taxa (identified to family level) from Albertine rift valley streams. Our findings showed that all five major functional feeding groups were represented; Gatherer Collectors (GC); Predators (PR); shredders (SH); Scrapers (SC); and Filterer collectors. The most dominant functional feeding group was the Gatherer Collectors (GC) that accounted for 53.5% of the total population. The most abundant (GC) families were Baetidae (7813 individuals), Chironomidae NTP (5628) and Caenidae (1848). Majority of the macroinvertebrate population feed on Fine particulate organic matter (FPOM) from the stream bottom. In terms of taxa richness the Predators (PR) had the highest value of 24 taxa and the Filterer Collectors group had the least number of taxa (3). The families that had the highest number of predators (PR) were Corixidae (1024 individuals), Coenagrionidae (445) and Libellulidae (283). However, Predators accounted for only 7.4% of the population. The findings highlighted the functional feeding groups and habitat type of macroinvertebrate communities along an altitudinal gradient.

Keywords: trophic levels, functional feeding groups, macroinvertebrates, Albertine rift

Procedia PDF Downloads 235

Authors: Jinan Fiaidhi, Sabah Mohammed, Petros Zezos

Abstract:

Capsule endoscopy (CE) is an established noninvasive diagnostic modality in investigating small bowel disease. CE has a pivotal role in assessing patients with suspected bleeding or identifying evidence of active Crohn's disease in the small bowel. However, CE produces lengthy videos with at least eighty thousand frames, with a frequency rate of 2 frames per second. Gastroenterologists cannot dedicate 8 to 15 hours to reading the CE video frames to arrive at a diagnosis. This is why the issue of analyzing CE videos based on modern artificial intelligence techniques becomes a necessity. However, machine learning, including deep learning, has failed to report robust results because of the lack of large samples to train its neural nets. In this paper, we are describing a thick data approach that learns from a few anchor images. We are using sound datasets like KVASIR and CrohnIPI to filter candidate frames that include interesting anomalies in any CE video. We are identifying candidate frames based on feature extraction to provide representative measures of the anomaly, like the size of the anomaly and the color contrast compared to the image background, and later feed these features to a decision tree that can classify the candidate frames as having a condition like the Crohn's Disease. Our thick data approach reported accuracy of detecting Crohn's Disease based on the availability of ulcer areas at the candidate frames for KVASIR was 89.9% and for the CrohnIPI was 83.3%. We are continuing our research to fine-tune our approach by adding more thick data methods for enhancing diagnosis accuracy.

Keywords: thick data analytics, capsule endoscopy, Crohn’s disease, siamese neural network, decision tree

Procedia PDF Downloads 156

Authors: Masoud Irannezhad, Sirpa Rasmus, Saghar Ahmadian, Deliang Chen, Bjorn Klove

Abstract:

Variability and changes in thermal conditions play a crucial role in functioning of human society, particularly over cold climate regions like Finland. Accordingly, the frost-free season (FFS) parameters in terms of start (FFSS), end (FFSE) and length (FFSL) have substantial effects not only on natural environment (e.g. flora and fauna), but also on human requirements (e.g. agriculture, forestry and energy generation). Applying the 0°C threshold of minimum temperature (Tmin), the FFS was defined as the period between the last spring frost as FFSS and the first fall frost as FFSE. For this study, gridded (10 x 10 km2) daily minimum temperature datasets throughout Finland during 1961-2011 was used to investigate recent spatio-temporal variations and trends in frost-free season (FFS) parameters and their relationships with the well-known large-scale climate teleconnections (CTs). The FFS in Finland naturally increases from north (~60 days) to south (~190 days), in association with earlier FFSS (~24 April) and later FFSE (~30 October). Statistically significant (p<0.05) trends in FFSL were all positive (increasing) ranged between 0 and 13.5 (days/decade) and mainly observed in the east, upper west, centre and upper north of Finland. Such lengthening trends in FFS were attributable to both earlier FFSS and later FFSE mostly over central and upper northern Finland, while only to later FFSE in eastern and upper western parts. Variations in both FFSL and FFSS were significantly associated with the Polar (POL) pattern over northern Finland, while with the East Atlantic (EA) pattern over eastern and upper western areas. However, the POL and Scandinavia (SCA) patterns were most influential CTs for FFSE variability over northern Finland.

Keywords: climate teleconnections, Finland, frost-free season, trend analysis

Procedia PDF Downloads 204

Authors: Emmanuel Okine Neokye Serge Dossou Martin Iniga Bortey Nketia Alabi-Doku

Abstract:

Fish stock distribution is greatly influenced by oceanographic environmental conditions. Temporal variations of temperature and other oceanic properties, resulting from climate change have been documented to have a strong impact on fisheries and aquaculture. In Ghana, Sardinella species are one of the most important fisheries resources; they constitute about 60% of the total catch of coastal fisheries and are more predominant during the upwelling season. The present study investigated the role of physical oceanographic environmental conditions in the catches of Sardinella species: S. aurita and S. maderensis, which were landed in Ghana. Furthermore, we examined the relationship between environmental conditions and catches of Sardinella species for seasonal and interannual variations between 2005 and 2015. For oceanographic environmental factors, we used comprehensive datasets, which consist of :(1) daily in situ SST data obtained at two coastal stations in Ghana; (i) Cape 3 Points (4.7° N, -2.09° W) and (ii) Tema (5° N, 0° E), for the period 2005–2015, (2) Monthly SST data (MOAA GPV) from JAMSTEC, and (3) gridded 10 metre wind data from CCMP reanalysis. The analysis of the data collected showed that higher (lower) wind velocity forms stronger (weaker) coastal upwelling that is detected by lower (higher) SST, resulting in a higher (lower) catch of Sardinella spp., in both seasonal and interannual variations. It was also observed that the capture ability of small pelagic fish species such as Sardinella spp. is depend on the intensity of the coastal upwelling. Moreso, the Atlantic Meridional Mode index (climatic index) is now known to be a possible factor to the interannual variation in catch of small pelagic fish species.

Keywords: Sardinella spp., fish, climate change, Ghana

Procedia PDF Downloads 15

Authors: Ahmed M. Khedr, Saad M. Abdelrahman, Kareem M. Tonbol

Abstract:

Two geo-referenced sea level datasets (September 2008 – November 2010) and (April 2012 – January 2014) were recorded at Alexandria Western Harbour (AWH). Accurate re-definition of tidal datum, referred to the latest International Terrestrial Reference Frame (ITRF-2014), was discussed and updated to improve our understanding of the old predefined tidal datum at Alexandria. Tidal and non-tidal components of sea level were separated with the use of Delft-3D hydrodynamic model-tide suit (Delft-3D, 2015). Tidal characteristics at AWH were investigated and harmonic analysis showed the most significant 34 constituents with their amplitudes and phases. Tide was identified as semi-diurnal pattern as indicated by a “Form Factor” of 0.24 and 0.25, respectively. Principle tidal datums related to major tidal phenomena were recalculated referred to a meaningful geodetic height datum. The portion of residual energy (surge) out of the total sea level energy was computed for each dataset and found 77% and 72%, respectively. Power spectral density (PSD) showed accurate resolvability in high band (1–6) cycle/days for the nominated independent constituents, except some neighbouring constituents, which are too close in frequency. Wind and atmospheric pressure data, during the recorded sea level time, were analysed and cross-correlated with the surge signals. Moderate association between surge and wind and atmospheric pressure data were obtained. In addition, long-term sea level rise trend at AWH was computed and showed good agreement with earlier estimated rates.

Keywords: Alexandria, Delft-3D, Egypt, geodetic reference, harmonic analysis, sea level

Procedia PDF Downloads 166

Authors: Abdul Khalique Baloch, Ali Okatan

Abstract:

The quality analysis of food and vegetable from image is hot topic now a day, where researchers make them better then pervious findings through different technique and methods. In this research we have review the literature, and find gape from them, and suggest better proposed approach, design the algorithm, developed a software to measure the quality from images, where accuracy of image show better results, and compare the results with Perouse work done so for. The Application we uses an open-source dataset and python language with tensor flow lite framework. In this research we focus to sort food and vegetable from image, in the images, the application can sorts and make them grading after process the images, it could create less errors them human base sorting errors by manual grading. Digital pictures datasets were created. The collected images arranged by classes. The classification accuracy of the system was about 94%. As fruits and vegetables play main role in day-to-day life, the quality of fruits and vegetables is necessary in evaluating agricultural produce, the customer always buy good quality fruits and vegetables. This document is about quality detection of fruit and vegetables using images. Most of customers suffering due to unhealthy foods and vegetables by suppliers, so there is no proper quality measurement level followed by hotel managements. it have developed software to measure the quality of the fruits and vegetables by using images, it will tell you how is your fruits and vegetables are fresh or rotten. Some algorithms reviewed in this thesis including digital images, ResNet, VGG16, CNN and Transfer Learning grading feature extraction. This application used an open source dataset of images and language used python, and designs a framework of system.

Keywords: deep learning, computer vision, image processing, rotten fruit detection, fruits quality criteria, vegetables quality criteria

Procedia PDF Downloads 70

Authors: Annalakshmi G., Sakthivel Murugan S.

Abstract:

This paper presents a local directional encoded derivative binary pattern (LDEDBP) feature extraction method that can be applied for the classification of submarine coral reef images. The classification of coral reef images using texture features is difficult due to the dissimilarities in class samples. In coral reef image classification, texture features are extracted using the proposed method called local directional encoded derivative binary pattern (LDEDBP). The proposed approach extracts the complete structural arrangement of the local region using local binary batten (LBP) and also extracts the edge information using local directional pattern (LDP) from the edge response available in a particular region, thereby achieving extra discriminative feature value. Typically the LDP extracts the edge details in all eight directions. The process of integrating edge responses along with the local binary pattern achieves a more robust texture descriptor than the other descriptors used in texture feature extraction methods. Finally, the proposed technique is applied to an extreme learning machine (ELM) method with a meta-heuristic algorithm known as weighted distance grey wolf optimizer (GWO) to optimize the input weight and biases of single-hidden-layer feed-forward neural networks (SLFN). In the empirical results, ELM-WDGWO demonstrated their better performance in terms of accuracy on all coral datasets, namely RSMAS, EILAT, EILAT2, and MLC, compared with other state-of-the-art algorithms. The proposed method achieves the highest overall classification accuracy of 94% compared to the other state of art methods.

Keywords: feature extraction, local directional pattern, ELM classifier, GWO optimization

Procedia PDF Downloads 164

Authors: Hadjer Sadoune, Liza Lamini, Scherazade Krim, Amel Djouadi, Rachida Rihani

Abstract:

Deep neural networks are highly regarded for their accuracy in predicting intricate fermentation processes. Their ability to learn from a large amount of datasets through artificial intelligence makes them particularly effective models. The primary obstacle in improving the performance of these models is to carefully choose the suitable hyperparameters, including the neural network architecture (number of hidden layers and hidden units), activation function, optimizer, learning rate, and other relevant factors. This study predicts biogas production from real wastewater treatment plant data using a sophisticated approach: hybrid Bayesian optimization with a tree-structured Parzen estimator (BO-TPE) for an optimised deep neural network (DNN) model. The plant utilizes an Upflow Anaerobic Sludge Blanket (UASB) digester that treats industrial wastewater from soft drinks and breweries. The digester has a working volume of 1574 m3 and a total volume of 1914 m3. Its internal diameter and height were 19 and 7.14 m, respectively. The data preprocessing was conducted with meticulous attention to preserving data quality while avoiding data reduction. Three normalization techniques were applied to the pre-processed data (MinMaxScaler, RobustScaler and StandardScaler) and compared with the Non-Normalized data. The RobustScaler approach has strong predictive ability for estimating the volume of biogas produced. The highest predicted biogas volume was 2236.105 Nm³/d, with coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE) values of 0.712, 164.610, and 223.429, respectively.

Keywords: anaerobic digestion, biogas production, deep neural network, hybrid bo-tpe, hyperparameters tuning

Procedia PDF Downloads 39

Authors: Wenjiang Deng, Tian Mou, Yudi Pawitan, Trung Nghia Vu

Abstract:

Estimation of isoform-level gene expression from RNA-seq data depends on simplifying assumptions, such as uniform reads distribution, that are easily violated in real data. Such violations typically lead to biased estimates. Most existing methods provide a bias correction step(s), which is based on biological considerations, such as GC content–and applied in single samples separately. The main problem is that not all biases are known. For example, new technologies such as single-cell RNA-seq (scRNA-seq) may introduce new sources of bias not seen in bulk-cell data. This study introduces a method called XAEM based on a more flexible and robust statistical model. Existing methods are essentially based on a linear model Xβ, where the design matrix X is known and derived based on the simplifying assumptions. In contrast, XAEM considers Xβ as a bilinear model with both X and β unknown. Joint estimation of X and β is made possible by simultaneous analysis of multi-sample RNA-seq data. Compared to existing methods, XAEM automatically performs empirical correction of potentially unknown biases. XAEM implements an alternating expectation-maximization (AEM) algorithm, alternating between estimation of X and β. For speed XAEM utilizes quasi-mapping for read alignment, thus leading to a fast algorithm. Overall XAEM performs favorably compared to other recent advanced methods. For simulated datasets, XAEM obtains higher accuracy for multiple-isoform genes, particularly for paralogs. In a differential-expression analysis of a real scRNA-seq dataset, XAEM achieves substantially greater rediscovery rates in an independent validation set.

Keywords: alternating EM algorithm, bias correction, bilinear model, gene expression, RNA-seq

Procedia PDF Downloads 143

Authors: M. Bilal Ishfaq, Adnan N. Qureshi

Abstract:

COVID-19 is the most infectious disease these days, it was first reported in Wuhan, the capital city of Hubei in China then it spread rapidly throughout the whole world. Later on 11 March 2020, the World Health Organisation (WHO) declared it a pandemic. Since COVID-19 is highly contagious, it has affected approximately 219M people worldwide and caused 4.55M deaths. It has brought the importance of accurate diagnosis of respiratory diseases such as pneumonia and COVID-19 to the forefront. In this paper, we propose a hybrid approach for the automated detection of COVID-19 using medical imaging. We have presented the hybridization of manually extracted and convolutional features. Our approach combines Haralick texture features and convolutional features extracted from chest X-rays and CT scans. We also employ a minimum redundancy maximum relevance (MRMR) feature selection algorithm to reduce computational complexity and enhance classification performance. The proposed model is evaluated on four publicly available datasets, including Chest X-ray Pneumonia, COVID-19 Pneumonia, COVID-19 CTMaster, and VinBig data. The results demonstrate high accuracy and effectiveness, with 0.9925 on the Chest X-ray pneumonia dataset, 0.9895 on the COVID-19, Pneumonia and Normal Chest X-ray dataset, 0.9806 on the Covid CTMaster dataset, and 0.9398 on the VinBig dataset. We further evaluate the effectiveness of the proposed model using ROC curves, where the AUC for the best-performing model reaches 0.96. Our proposed model provides a promising tool for the early detection and accurate diagnosis of COVID-19, which can assist healthcare professionals in making informed treatment decisions and improving patient outcomes. The results of the proposed model are quite plausible and the system can be deployed in a clinical or research setting to assist in the diagnosis of COVID-19.

Keywords: COVID-19, feature engineering, artificial neural networks, radiology images

Procedia PDF Downloads 76

Authors: Zahid Ullah, Atlas Khan

Abstract:

The rapid growth of data in various domains has created a pressing need for effective methods to transform this data into meaningful knowledge. In this era of big data, mathematical and statistical innovations play a crucial role in unlocking insights and facilitating informed decision-making in data analytics. This abstract aims to explore the transformative potential of these innovations and their impact on converting raw data into actionable knowledge. Drawing upon a comprehensive review of existing literature, this research investigates the cutting-edge mathematical and statistical techniques that enable the conversion of data into knowledge. By evaluating their underlying principles, strengths, and limitations, we aim to identify the most promising innovations in data analytics. To demonstrate the practical applications of these innovations, real-world datasets will be utilized through case studies or simulations. This empirical approach will showcase how mathematical and statistical innovations can extract patterns, trends, and insights from complex data, enabling evidence-based decision-making across diverse domains. Furthermore, a comparative analysis will be conducted to assess the performance, scalability, interpretability, and adaptability of different innovations. By benchmarking against established techniques, we aim to validate the effectiveness and superiority of the proposed mathematical and statistical innovations in data analytics. Ethical considerations surrounding data analytics, such as privacy, security, bias, and fairness, will be addressed throughout the research. Guidelines and best practices will be developed to ensure the responsible and ethical use of mathematical and statistical innovations in data analytics. The expected contributions of this research include advancements in mathematical and statistical sciences, improved data analysis techniques, enhanced decision-making processes, and practical implications for industries and policymakers. The outcomes will guide the adoption and implementation of mathematical and statistical innovations, empowering stakeholders to transform data into actionable knowledge and drive meaningful outcomes.

Keywords: data analytics, mathematical innovations, knowledge extraction, decision-making

Procedia PDF Downloads 75

Authors: Fredrick Etyang

Abstract:

This paper presents a Rapid Industrialization Model (RIM) designed to support existing industrialization policies, strategies and industrial development plans at National, Regional and Constituent level in Africa. The model will reinforce efforts to attainment of inclusive and sustainable industrialization of Africa by state and non-state actors. The overall objective of this model is to serve as a framework for rapid industrialization in developing economies and the specific objectives range from supporting rapid industrialization development to promoting a structural change in the economy, a balanced regional industrial growth, achievement of local, regional and international competitiveness in areas of clear comparative advantage in industrial exports and ultimately, the RIM will serve as a step-by-step guideline for the industrialization of African Economies. This model is a product of a scientific research process underpinned by desk research through the review of African countries development plans, strategies, datasets, industrialization efforts and consultation with key informants. The rigorous research process unearthed multi-directional and renewed efforts towards industrialization of Africa premised on collective commitment of individual states, regional economic communities and the African union commission among other strategic stakeholders. It was further, established that the inputs into industrialization of Africa outshine the levels of industrial development on the continent. The RIM comes in handy to serve as step-by-step framework for African countries to follow in their industrial development efforts of transforming inputs into tangible outputs and outcomes in the short, intermediate and long-run. This model postulates three stages of industrialization and three phases toward rapid industrialization of African economies, the model is simple to understand, easily implementable and contextualizable with high return on investment for each unit invested into industrialization supported by the model. Therefore, effective implementation of the model will result into inclusive and sustainable rapid industrialization of Africa.

Keywords: economic development, industrialization, economic efficiency, exports and imports

Procedia PDF Downloads 85

Authors: Susan Diamond

Abstract:

Machine Learning workloads have traditionally been run in high-performance computing (HPC) environments, where users log in to dedicated machines and utilize the attached GPUs to run training jobs on huge datasets. Training of large neural network models is very resource intensive, and even after exploiting parallelism and accelerators such as GPUs, a single training job can still take days. Consequently, the cost of hardware is a barrier to entry. Even when upfront cost is not a concern, the lead time to set up such an HPC environment takes months from acquiring hardware to set up the hardware with the right set of firmware, software installed and configured. Furthermore, scalability is hard to achieve in a rigid traditional lab environment. Therefore, it is slow to react to the dynamic change in the artificial intelligent industry. Watson Deep Learning as a service, a cloud-based deep learning platform that mitigates the long lead time and high upfront investment in hardware. It enables robust and scalable sharing of resources among the teams in an organization. It is designed for on-demand cloud environments. Providing a similar user experience in a multi-tenant cloud environment comes with its own unique challenges regarding fault tolerance, performance, and security. Watson Deep Learning as a service tackles these challenges and present a deep learning stack for the cloud environments in a secure, scalable and fault-tolerant manner. It supports a wide range of deep-learning frameworks such as Tensorflow, PyTorch, Caffe, Torch, Theano, and MXNet etc. These frameworks reduce the effort and skillset required to design, train, and use deep learning models. Deep Learning as a service is used at IBM by AI researchers in areas including machine translation, computer vision, and healthcare. 

Keywords: deep learning, machine learning, cognitive computing, model training

Procedia PDF Downloads 209

Authors: Paul Longley Arthur

Abstract:

This paper reports on an Australian Research Council-funded project utilising the Australian digital research infrastructure the ‘Time-Layered Cultural Map of Australia’ (TLCMap) (https://www.tlcmap.org/) [1]. This resource has been developed to help researchers create digital maps from cultural, textual, and historical data, layered with datasets registered on the platform. TLCMap is a set of online tools that allows humanities researchers to compile humanities data using spatio-temporal coordinates – to upload, gather, analyse and visualise data. It is the only purpose-designed, Australian-developed research tool for humanities and social science researchers to identify geographical clusters and parallel journeys by sight. This presentation discusses a series of Aboriginal mapping and visualisation experiments using TLCMap to show how Indigenous knowledge can reconfigure contemporary understandings of space including the urbanised landscape [2, 3]. The research data being generated – investigating the historical movements of Aboriginal people, the distribution of networks, and their relation to land – lends itself to mapping and geo-spatial visualisation and analysis. TLCMap allows researchers to create layers on a 3D map which pinpoint locations with accompanying information, and this has enabled our research team to plot out traditional historical journeys undertaken by Aboriginal people as well as to compile a gazetteer of Aboriginal place names, many of which have largely been undocumented until now [4]. The documented journeys intersect with and overlay many of today’s urban formations including main roads, municipal boundaries, and state borders. The paper questions how such data can be incorporated into a more culturally and ethically responsive understanding of contemporary urban spaces and as well as natural environments [5].

Keywords: spatio-temporal mapping, visualisation, Indigenous knowledge, mobility and migration, research infrastructure

Procedia PDF Downloads 21

Authors: Babak Bahri, Fatemeh Yassaee Meybodi, Changiz Eslahchi

Abstract:

In the pursuit of effective cancer therapies, the exploration of combinatorial drug regimens is crucial to leverage synergistic interactions between drugs, thereby improving treatment efficacy and overcoming drug resistance. However, identifying synergistic drug pairs poses challenges due to the vast combinatorial space and limitations of experimental approaches. This study introduces ClusterSyn, a machine learning (ML)-powered framework for classifying anti-cancer drug synergy scores. ClusterSyn employs a two-step approach involving drug clustering and synergy score prediction using a fully connected deep neural network. For each cell line in the training dataset, a drug graph is constructed, with nodes representing drugs and edge weights denoting synergy scores between drug pairs. Drugs are clustered using the Markov clustering (MCL) algorithm, and vectors representing the similarity of drug pairs to each cluster are input into the deep neural network for synergy score prediction (synergy or antagonism). Clustering results demonstrate effective grouping of drugs based on synergy scores, aligning similar synergy profiles. Subsequently, neural network predictions and synergy scores of the two drugs on others within their clusters are used to predict the synergy score of the considered drug pair. This approach facilitates comparative analysis with clustering and regression-based methods, revealing the superior performance of ClusterSyn over state-of-the-art methods like DeepSynergy and DeepDDS on diverse datasets such as Oniel and Almanac. The results highlight the remarkable potential of ClusterSyn as a versatile tool for predicting anti-cancer drug synergy scores.

Keywords: drug synergy, clustering, prediction, machine learning., deep learning

Procedia PDF Downloads 81

Authors: Abdullah Al Farwan, Ya Zhang

Abstract:

In today’s educational arena, it is critical to understand educational data and be able to evaluate important aspects, particularly data on student achievement. Educational Data Mining (EDM) is a research area that focusing on uncovering patterns and information in data from educational institutions. Teachers, if they are able to predict their students' class performance, can use this information to improve their teaching abilities. It has evolved into valuable knowledge that can be used for a wide range of objectives; for example, a strategic plan can be used to generate high-quality education. Based on previous data, this paper recommends employing data mining techniques to forecast students' final grades. In this study, five data mining methods, Decision Tree, JRip, Naive Bayes, Multi-layer Perceptron, and Random Forest with wrapper feature selection, were used on two datasets relating to Portuguese language and mathematics classes lessons. The results showed the effectiveness of using data mining learning methodologies in predicting student academic success. The classification accuracy achieved with selected algorithms lies in the range of 80-94%. Among all the selected classification algorithms, the lowest accuracy is achieved by the Multi-layer Perceptron algorithm, which is close to 70.45%, and the highest accuracy is achieved by the Random Forest algorithm, which is close to 94.10%. This proposed work can assist educational administrators to identify poor performing students at an early stage and perhaps implement motivational interventions to improve their academic success and prevent educational dropout.

Keywords: classification algorithms, decision tree, feature selection, multi-layer perceptron, Naïve Bayes, random forest, students’ academic performance

Procedia PDF Downloads 169

Authors: Makhosi Buthelezi

Abstract:

Mitochondrial DNA cytochrome c oxidase I (COI) gene analyses linked the South African groundnut leaf miner (GLM) to the Australian soya bean moth Aproaerema simplexella (Walker) and Indian Aproaerema modicella (Deventer). Thus, the genetic relatedness of GLM, A. simplexela, and A. modicella was examined by performing mitochondrial and nuclear (COI, cytochrome oxidase subunit II (COII), mitochondrial cytochrome b (CYTB), nuclear ribosomal 28S (28S) and intergenic spacer elongation factor-1 alpha ( EF-1 ALPHA) on 44 specimens collected from South Africa, four from Mozambique, and three each from single locations in India and Australia. Phylogenetic analyses were conducted using the Maximum Parsimony (MP) and Neighbour-Joining (NJ) methods. All of the datasets of the five DNA gene regions that were sequenced were also analyzed using the Basic Local Alignment Search Tool (BLAST) to find the closest matches for inclusion in the phylogenetic trees as outgroups and for purposes of information. In the phylogenetic trees for COI, COII, cytb and EF-1 ALPHA, a similar pattern was observed in the way that the sequences assembled into different groups; i.e., some sequences of A. simplexella from Australia were grouped separately from the others, but some Australian sequences grouped with those of the GLM from South Africa, India, and Mozambique. In the phylogenetic tree for 28S, all sequences from South Africa, Australia, India, and Mozambique grouped together and formed one group. For COI, genetic pairwise distance ranged from 0.97 to 3.60 %, for COII it ranged from 0.19% to 2.32%, for cytb it ranged from 0.25 to 9.77% and for EF-1 ALPHA it ranged 0.48 to 6.99%. Results of this study indicate that these populations are genetically related and presumably constitute a single species. Thus, further molecular and morphological studies need to be undertaken in order to resolve this apparent conundrum on the taxonomy of these populations.

Keywords: aproaerema modicella, aproaerema simplexella, mitochondrial DNA, nuclear DNA

Procedia PDF Downloads 200

Authors: Aminu Abdullahi Isyaku

Abstract:

Semi-arid Bornu basin in northeast Nigeria is characterised with flat topography, thick cover sediments and lack of continuous bedrock outcrops discernible for field geology. This paper presents the methodology for the characterisation of neotectonic surface structures and surface lithology in the north-eastern Bornu basin in northeast Nigeria as an alternative approach to field geological mapping using free multispectral Landsat 7 ETM+, SRTM DEM and ASAR Earth Observation datasets. Spectral lithological mapping herein developed utilised spectral discrimination of the surface features identified on Landsat 7 ETM+ images to infer on the lithology using four steps including; computations of band combination images; band ratio images; supervised image classification and inferences of the lithological compositions. Two complementary approaches to lineament mapping are carried out in this study involving manual digitization and automatic lineament extraction to validate the structural lineaments extracted from the Landsat 7 ETM+ image mosaic covering the study. A comparison between the mapped surface lineaments and lineament zones show good geospatial correlation and identified the predominant NE-SW and NW-SE structural trends in the basin. Topographic profiles across different parts of the Bama Beach Ridge palaeoshorelines in the basin appear to show different elevations across the feature. It is determined that most of the drainage systems in the northeastern Bornu basin are structurally controlled with drainage lines terminating against the paleo-lake border and emptying into the Lake Chad mainly arising from the extensive topographic high-stand Bama Beach Ridge palaeoshoreline.

Keywords: Bornu Basin, lineaments, spectral lithology, tectonics

Procedia PDF Downloads 142

Authors: Paul Rowley, Adam Thirkill, Nick Doylend, Philip Leicester, Becky Gough

Abstract:

The dynamic performance of all energy generation technologies is impacted to varying degrees by the stochastic properties of the wider system within which the generation technology is located. This stochasticity can include the varying nature of ambient renewable energy resources such as wind or solar radiation, or unpredicted changes in energy demand which impact upon the operational behaviour of thermal generation technologies. An understanding of these stochastic impacts are especially important in contexts such as highly distributed (or embedded) generation, where an understanding of issues affecting the individual or aggregated performance of high numbers of relatively small generators is especially important, such as in ESCO projects. Probabilistic evaluation of monitored or simulated performance data is one technique which can provide an insight into the dynamic performance characteristics of generating systems, both in a prognostic sense (such as the prediction of future performance at the project’s design stage) as well as in a diagnostic sense (such as in the real-time analysis of underperforming systems). In this work, we describe the development, application and outcomes of a new approach to the acquisition of datasets suitable for use in the subsequent performance and impact analysis (including the use of Bayesian approaches) for a number of distributed generation technologies. The application of the approach is illustrated using a number of case studies involving domestic and small commercial scale photovoltaic, solar thermal and natural gas boiler installations, and the results as presented show that the methodology offers significant advantages in terms of plant efficiency prediction or diagnosis, along with allied environmental and social impacts such as greenhouse gas emission reduction or fuel affordability.

Keywords: renewable energy, dynamic performance simulation, Bayesian analysis, distributed generation

Procedia PDF Downloads 495

Authors: Victoria Masi Haruna Karatu, Nik Kamariah Nikmat

Abstract:

In recent times awareness about the environment and green purchase has been on the increase across nations due to global warming. Previous researchers have attempted to determine what actually influences the purchase intention of consumers in this environmentally conscious epoch. The consumers too have become conscious of what to buy and who to buy from in their purchasing decisions as this action will reflect their concern about the environment and their personal well-being. This trend is a widespread phenomenon in most developed countries of the world. On the contrary evidence revealed that only 5% of the populations of Nigeria involve in green purchase activities thus making the country lag behind its counterparts in green practices. This is not a surprise as Nigeria is facing problems of inadequate green knowledge, non-enforcement of environmental regulations, sensitivity to the price of green products when compared with the conventional ones and distrust towards green products which has been deduced from prior studies of other regions. The main objectives of this study is to examine the direct antecedents of green purchase intention (green availability, government regulations, perceived green knowledge, perceived value and green price sensitivity) in Nigeria and secondly to establish the mediating role of perceived behavioral control on the relationship between these antecedents and green purchase intention. The study adopts quantitative method whereby 700 questionnaires were administered to lecturers in three Nigerian universities. 502 datasets were collected which represents 72 percent response rate. After screening the data only 440 were usable and analyzed using structural equation modeling (SEM) and bootstrapping. From the findings, three antecedents have significant direct relationships with green purchase intention (perceived green knowledge, perceived behavioral control, and green availability) while two antecedents have positive and significant direct relationship with perceived behavioral control (perceived value and green price sensitivity). On the other hand, PBC does not mediate any of the paths from the predictors to criterion variable. This result is discussed in the Nigerian context.

Keywords: Green Availability, Green Price Sensitivity, Green Purchase Intention, Perceived Green Knowledge, Perceived Value

Procedia PDF Downloads 427

Authors: E. N. Masri, E. Takács

Abstract:

Amplitude Versus Offset (AVO) and simultaneous model-based impedance inversion techniques have not been utilized for geothermal exploration commonly; however, some recent publications called the attention that they can be very useful in the geothermal investigations. In this study, we present rock physical attributes obtained from 3D pre-stack seismic data and well logs collected in a study area of the NW part of Pannonian Basin where the geothermal reservoir is located in the fractured zones of Triassic basement and it was hit by three productive-injection well pairs. The holes were planned very successfully based on the conventional 3D migrated stack volume prior to this study. Subsequently, the available geophysical-geological datasets provided a great opportunity to test modern inversion procedures in the same area. In this presentation, we provide a summary of the theory and application of the most promising seismic inversion techniques from the viewpoint of geothermal exploration. We demonstrate P- and S-wave impedance, as well as the velocity (Vp and Vs), the density, and the Vp/Vs ratio attribute volumes calculated from the seismic and well-logging data sets. After a detailed discussion, we conclude that P-wave impedance and Vp/Vp ratio are the most helpful parameters for lithology discrimination in the study area. They detect the hot water saturated fracture zone very well thus they can be very useful in mapping the investigated reservoir. Integrated interpretation of all the obtained rock-physical parameters is essential. We are extending the above discussed pre-stack seismic tools by studying the possibilities of Elastic Impedance Inversion (EII) for geothermal exploration. That procedure provides two other useful rock-physical properties, the compressibility and the rigidity (Lamé parameters). Results of those newly created elastic parameters will also be demonstrated in the presentation. Geothermal extraction is of great interest nowadays; and we can adopt several methods have been successfully applied in the hydrocarbon exploration for decades to discover new reservoirs and reduce drilling risk and cost.

Keywords: fractured zone, seismic, well-logging, inversion

Procedia PDF Downloads 128

Authors: Chloé Kernaléguen, Laura Kundun, Tessie Lery, Ryan Laleg, Zhangyun Tan

Abstract:

This study explores global disparities in respiratory syncytial virus (RSV) bronchiolitis incidence among children aged 0-2 years, focusing on the human development index (HDI) as a key determinant. RSV bronchiolitis poses a significant health risk to young children, influenced by factors, including socio-economic conditions captured by the HDI. Through a comprehensive systematic review and dataset selection (Switzerland, Brazil, United States of America), we formulated an HDI-SEIRS numerical model within the SEIRS framework. Results show variations in RSV bronchiolitis dynamics across countries, emphasizing the influence of HDI. Modelling reveals a correlation between higher HDI and increased bronchiolitis spread, notably in the USA and Switzerland. The ratios HDIcountry over HDImax strengthen this association, while climate disparities contribute to variations, especially in colder climates like the USA and Switzerland. The study raises the hypothesis of an indirect link between higher HDI and more frequent bronchiolitis, underlining the need for nuanced understanding. Factors like improved healthcare access, population density, mobility, and social behaviors in higher HDI countries might contribute to unexpected trends. Limitations include dataset quality and restricted RSV bronchiolitis data. Future research should encompass diverse HDI datasets to refine HDI's role in bronchiolitis dynamics. In conclusion, HDI-SEIRS models offer insights into factors influencing RSV bronchiolitis spread. While HDI is a significant indicator, its impact is indirect, necessitating a holistic approach to effective public health policies. This analysis sets the stage for further investigations into multifaceted interactions shaping bronchiolitis dynamics in diverse socio-economic contexts.

Keywords: bronchiolitis propagation, HDI influence, respiratory syncytial virus, SEIRS model

Procedia PDF Downloads 68

Authors: Siqin Wang, Jonathan Corcoran, Yan Liu, Thomas Sigler

Abstract:

Australia is a quintessentially immigrant nation with 28 percent of its residents being foreign-born. By 2011, People’s Republic of China (PRC) overtook the United Kingdom to become the largest source country in Australia. Significantly, the profile of PRC-born migrants has changed to mirror broader global shifts towards high-skilled labour, education-related, and investment-focussed migration, all of which reflect an increasing trend in the mobility of wealthy and/or educated cohorts. Together, these coalesce to form a more complex pattern of migrant settlement –both spatially and socio-economically. This paper focuses on the PRC-born migration, redresses these lacunae, with regard to the settlement outcomes of PRC migrants to Australia, with a particular focus on spatial evolution and residential mobility at both the metropolitan and national scales. By drawing on Census Data and migration Micro Datasets, the aim of this paper is to examine the shifting dynamics of PRC-born migrants in Australian capital cities to unveil their socioeconomic characteristics, residential patterns and change of spatial concentrations during their transition into the new host society. This paper finds out three general patterns in the residential evolution of PRC-born migrants depending on the size of capital cities where they settle down, as well as the association of socio-economic characters with the formation of enclaves. It also examines the residential mobility across states and cities from 2001 to 2011 indicating the rising status of median-size Australian capital cities for receiving PRC-born migrants. The paper concludes with a discussion of evidences for policy formation, facilitates the effective transition of PRC-born populations into the mainstream of host society and enhances social harmony to help Australia become a more successful multicultural nation.

Keywords: Australia, Chinese migrants, residential mobility, spatial evolution

Procedia PDF Downloads 233

Authors: Una Carthy

Abstract:

A recent study of third level institutions in Ireland reveals that both age and aptitude can be overcome by teaching methodologies to motivate second language learners. This PhD investigation gathered quantitative and qualitative data from 14 Institutes of Technology over a three years period from 2011 to 2014. The fundamental research question was to establish the impact of institutional language policy on attitudes towards language learning. However, other related issues around second language acquisition arose in the course of the investigation. Data were collected from both lectures and students, allowing interesting points of comparison to emerge from both datasets. Negative perceptions among lecturers regarding language provision were often associated with the view that language learning belongs to primary and secondary level and has no place in third level education. This perception was offset by substantial data showing positive attitudes towards adult language learning. Lenneberg’s Critical Age Theory postulated that the optimum age for learning a second language is before puberty. More recently, scholars have challenged this theory in their studies, revealing that mature learners can and do succeed at learning languages. With regard to aptitude, a preoccupation among lecturers regarding poor literacy skills among students emerged and was often associated with resistance to second language acquisition. This was offset by a preponderance of qualitative data from students highlighting the crucial role which teaching approaches play in the learning process. Interestingly, the data collected regarding learning disabilities reveals that, given the appropriate learning environments, individuals can be motivated to acquire second languages, and indeed succeed at learning them. These findings are in keeping with other recent studies regarding attitudes towards second language learning among students with learning disabilities. Both sets of findings reinforce the case for language policies in the Institute of Technology (IoTs). Supportive and positive learning environments can be created in third level institutions to motivate adult learners, thereby overcoming perceived obstacles relating to age and aptitude.

Keywords: age, aptitude, second language acquisition, teaching methodologies

Procedia PDF Downloads 124

Authors: Mounir Azzam, Valerie Graw, Andreas Rienow

Abstract:

The Syrian war, which commenced in 2011, has resulted in significant changes in the real estate market in the Damascus metropolitan area, with rising levels of insecurity and disputes over tenure rights. The quest for spatial justice is, therefore, imperative, and this study performs a spatiotemporal analysis to investigate the impact of the war on real estate differentiation. Using the hedonic price models including 2,411 housing transactions over the period 2010-2022, this study aims to understand the spatial dynamics of the real estate market in wartime. Our findings indicate that war variables have had a significant impact on the differentiation and depreciation of property prices. Notably, property attributes have a more substantial impact on real estate values than district location, with severely damaged buildings in Damascus city resulting in an 89% decline in prices, while prices in Rural Damascus districts have decreased by 50%. Additionally, this study examines the urban texture of Damascus using correlation and homogeneity statistics derived from the gray-level co-occurrence matrix obtained from Google Earth Engine. We monitored 250 samples from hedonic datasets within three different years of the Syrian war (2015, 2019, and 2022). Our findings show that correlation values were highly differentiated, particularly among Rural Damascus districts, with a total decline of 87.2%. While homogeneity values decreased overall between 2015 and 2019, they improved slightly after 2019. The findings have valuable implications, not only for investment prospects in setting up a successful reconstruction strategy but also for spatial justice of property rights in strongly encouraging sustainable real estate development.

Keywords: hedonic price, real estate differentiation, reconstruction strategy, spatial justice, urban texture analysis

Procedia PDF Downloads 89

Authors: Naomi W. Maina

Abstract:

Fertility preference is a subject of great significance in developing countries. Studies reveal that the preferences of fertility are actually significant in determining the society’s fertility levels because the fertility behavior of the future has a high likelihood of falling under the effect of currently observed fertility inclinations. The objective of this study was to establish the factors associated with fertility preference amongst couples in Kenya by fitting a multinomial logistic regression model against 5,265 couple data obtained from Kenya demographic health survey 2014. Results revealed that the type of place of residence, the region of residence, age and spousal age gap significantly influence desire for additional children among couples in Kenya. There was the notable high likelihood of couples living in rural settlements having similar fertility preference compared to those living in urban settlements. Moreover, geographical disparities such as in northern Kenya revealed significant differences in a couples desire to have additional children compared to Nairobi. The odds of a couple’s desire for additional children were further observed to vary dependent on either the wife or husbands age and to a large extent the spousal age gap. Evidenced from the study, was the fact that as spousal age gap increases, the desire for more children amongst couples decreases. Insights derived from this study would be attractive to demographers, health practitioners, policymakers, and non-governmental organizations implementing fertility related interventions in Kenya among other stakeholders. Moreover, with the adoption of devolution, there is a clear need for adoption of population policies that are County specific as opposed to a national population policy as is the current practice in Kenya. Additionally, researchers or students who have little understanding in the application of multinomial logistic regression, both theoretical understanding and practical analysis in SPSS as well as application on real datasets, will find this article useful.

Keywords: couples' desire, fertility, fertility preference, multinomial regression analysis

Procedia PDF Downloads 182

Authors: Sami Ayad, Mengshan Lee

Abstract:

The Sustainable Development Goals (SDGs) were designed to be the best possible blueprint to achieve peace, prosperity, and overall, a better and more sustainable future for the Earth and all its people, and such a blueprint is needed more than ever. The SDGs face many hurdles that will prevent them from becoming a reality, one of such hurdles, arguably, is the chemical pollution and unintended chemical impacts generated through the production of various goods and resources that we consume. Chemical Alternatives Assessment has proven to be a viable solution for chemical pollution management in terms of filtering out hazardous chemicals for a greener alternative. However, the current substitution practice lacks crucial quantitative datasets (exposures and life cycle impacts) to ensure no unintended trade-offs occur in the substitution process. A Chemical Life Cycle Alternative Assessment (CLiCAA) framework is proposed as a reliable and replicable alternative to Life Cycle Based Alternative Assessment (LCAA) as it integrates chemical molecular structure analysis and Chemical Life Cycle Collaborative (CLiCC) web-based tool to fill in data gaps that the former frameworks suffer from. The CLiCAA framework consists of a four filtering layers, the first two being mandatory, with the final two being optional assessment and data extrapolation steps. Each layer includes relevant impact categories of each chemical, ranging from human to environmental impacts, that will be assessed and aggregated into unique scores for overall comparable results, with little to no data. A feasibility study will demonstrate the efficiency and accuracy of CLiCAA whilst bridging both cancer potency and exposure limit data, hoping to provide the necessary categorical impact information for every firm possible, especially those disadvantaged in terms of research and resource management.

Keywords: chemical alternative assessment, LCA, LCAA, CLiCC, CLiCAA, chemical substitution framework, cancer potency data, chemical molecular structure analysis

Procedia PDF Downloads 92