Proposed Machine Learning Approach Research Articles

Walleye (Sander vitreus, Mitchill 1818) age and sex classification using innovative supervised and unsupervised machine learning and soft computing methodologies

Walleye (Sander vitreus) is a freshwater perciform native to Northern America and Canada with high commercial value. Stocking programs for the species are unable to supply the significant demand, creating stock management problems and an ecosystemic decline. For sustainable fisheries, the age and sex estimation are fundamental to marine stock management. Biological methods for age determination and sex estimation were used in the past, conventional observations and techniques are substituted with machine learning methods, currently underused. The aim of this study is to evaluate the performance of an ensemble of Machine Learning (ML) methods for age and sex classification of walleye. Based on the performance of the ensemble models, appropriate evaluation decides on the age criterion with near optimal performance using supervised and unsupervised ML algorithms on the given fish datasets. The innovation of this work lies in the fact that the species age or sex estimation is based on a paramount involvement of models that diminish the proprietary idiosyncrasies of individual methods on specific datasets. For the same models two approaches were used: applied model algorithms developed both in R and Python (Orange Rapid Application Development environment) programming language provided high accuracy percentages. Among all methodologies, the Decision Tree method provided the best classification, for age estimation with 96% and 98% accuracy using the R and Orange models respectively. For the sex classification, Random Forest performed the best using the R-model with 94% accuracy, outperformed by the AdaBoost model in Orange with 97.6% accuracy. When compared, the models which were developed in Python generally performed better than the corresponding R-models both in age and sex classification. Overall, the proposed ML approach is emphasized as a rapid economic solution to standard fish classification with high accuracy, and thus could be integrated in stock assessment for a more sustainable aquaculture management.

Vessel turnaround time prediction: A machine learning approach

Uncertainty in vessel turnaround time (VTT) is troublesome and would reduce the operational efficiency in port management, potentially causing economic losses. Despite vessels generally providing their estimated departure time (EDT), there is frequently a considerable difference between the EDT and the actual departure time (ADT) of vessels due to various factors such as unexpected port handling inefficiency. This variability complicates the coordination of efficient port operations. Our research aims to address this issue by employing an extreme gradient boosting (XGBoost) regression model to predict the VTT, using vessel arrival and departure data at the Hong Kong Port for the year 2022 and the first quarter of 2023. The proposed machine learning approach can provide more accurate predictions on VTT on average compared to the EDT data reported by vessels themselves, with a substantial reduction in both mean absolute error (MAE) and root mean square error (RMSE) of 23% (from 5.1 h to 3.9 h) and 24% (from 8.0 h to 6.1 h), respectively. These results present a significant leap forward in the predictive capabilities for the VTT and lay the foundation for further research into improving vessel scheduling efficiency, reducing port congestion and enhancing overall port performance.

Prediction of the Response of Masonry Walls under Blast Loading Using Artificial Neural Networks

A methodology to predict key aspects of the structural response of masonry walls under blast loading using artificial neural networks (ANN) is presented in this paper. The failure patterns of masonry walls due to in and out-of-plane loading are complex due to the potential opening and sliding of the mortar joint interfaces between the masonry stones. To capture this response, advanced computational models can be developed requiring a significant amount of resources and computational effort. The article uses an advanced non-linear finite element model to capture the failure response of masonry walls under blast loads, introducing unilateral contact-friction laws between stones and damage mechanics laws for the stones. Parametric finite simulations are automatically conducted using commercial finite element software linked with MATLAB R2019a and Python. A dataset is then created and used to train an artificial neural network. The trained neural network is able to predict the out-of-plane response of the masonry wall for random properties of the blast load (standoff distance and weight). The results indicate that the accuracy of the proposed framework is satisfactory. A comparison of the computational time needed for a single finite element simulation and for a prediction of the out-of-plane response of the wall by the trained neural network highlights the benefits of the proposed machine learning approach in terms of computational time and resources. Therefore, the proposed approach can be used to substitute time consuming explicit dynamic finite element simulations and used as a reliable tool in the fast prediction of the masonry response under blast actions.

Maximizing reusability of learning objects through machine learning techniques

Maximizing the reusability of learning objects through machine learning techniques has significantly transformed the landscape of e-learning systems. This progress has fostered authentic resource sharing and expanded opportunities for learners to explore these materials with ease. Consequently, a pressing need arises for an efficient categorization system to organize these learning objects effectively. This study consists of two primary phases. Firstly, we extract metadata from learning objects using web exploration algorithms, specifically employing feature selection techniques to identify the most relevant features while eliminating redundant ones. This step drastically reduces the dataset’s dimensionality, enabling the creation of practical and useful models. In the second phase, we employ machine learning algorithms to categorize learning objects based on their specific forms of similarity. These algorithms are adept at accurately classifying objects by measuring their similarity using Euclidean distance metrics. To evaluate the effectiveness of learning objects through machine learning techniques, a series of experimental studies were conducted using a real-world dataset. The results of this study demonstrate that the proposed machine learning approach surpasses traditional methods, yielding promising and efficient outcomes for enhancing learning object reusability.

Using machine learning to determine the positions of professional soccer players in terms of biomechanical variables

This study aimed to predict professional soccer players’ positions with machine learning according to certain locomotor demands. Data from 20 male professional soccer players (five defenders, eight midfielders, and seven attackers) from the same team were tracked daily with a global navigation satellite system. A total of 1910 individual training sessions were recorded. The 10-fold cross-validation method was used. Soccer player positions were predicted using predictive models created with random forest (RF), gradient boosting tree, bagging classification, and regression trees algorithms, and the results were evaluated with comprehensive performance measures. Ratios and an importance plot were used to analyze the importance of the variables according to their contributions to the estimation. The findings show that the RF model achieved 100% accuracy, which means that RF can predict all player positions (100%). Running distance (26.5%), total distance (17.2%), and player load (15.8%) were the three variables that contributed the most to the estimation of the RF model and were the most important factor in distinguishing player positions. Consequently, our proposed machine learning approach (RF model) can reduce false alarms and player mispositioning.

A machine learning tool for future prediction of heat release capacity of in-situ flame retardant hybrid Mg(OH)2-Epoxy nanocomposites

In this work, the fire behavior of a sol-gel in-situ hybrid Mg(OH)2-epoxy nanocomposite was investigated and an artificial neural network-based system built on a fully connected feed-forward artificial neural network was developed to predict its heat release capacity. The nanocomposite containing only ∼5 wt% loading of Mg(OH)2 promoted a remarkable decrease in heat release capacity (∼34%) measured by pyrolysis combustion flow calorimetry and in peak of heat release rate (∼37%), and heat release rate (∼29%), as assessed by cone calorimetry, as well as a significant decrease of total smoke release and smoke extinction area about 22 and 5%, respectively, indicating the suitability of Mg(OH)2 as an effective smoke suppressant. The proposed machine learning approach may be used as a promising alternative for a cost- and time-saving prediction of the fire performances of novel flame retardant polymer-based nanocomposites.

Vectorize Me! A Proposed Machine Learning Approach for Segmenting the Multi-optional Tourist

Contemporary consumer behavior is characterized by its multidimensionality and complexity, which, at the same time, pushes traditional segmentation approaches to their limits. In response, this methodological study proposes a multistage machine learning-based segmentation process using semiotic-semantic community detection. This innovative method was conducted exemplarily and evaluated on a representative sample of 1,101 German travelers. The main contribution of this study lies in the novel use of word vectors, which result from assigning a semiotic meaning to travel-type images. Thus, high-dimensional data could be used during the segmentation process, overcoming several classical segmentation problems. By using semantic similarities, tourists could be grouped and represented in their multidimensionality. From a theoretical perspective, this study was inspired by postmodern tourism practices in order to better understand the (oftentimes) hybrid and multilayered behaviors of tourists. To make this innovative approach reproducible, recommendations for implementation and all necessary data have been provided.

Machine learning simulation of one-dimensional deterministic water wave propagation

Deterministic phase-resolved prediction of the evolution of surface gravity waves in water is challenging due to their complex spatio-temporal dynamics. Physics-based methods of varying complexity are available, but the conflicting objectives of numerical efficiency and accuracy impede real-time wave prediction. Data-driven methods may be able to overcome this challenge by using training data generated by complex numerical methods. This work explores the potential of a machine learning (ML) approach based on a fully convolutional encoder–decoder architecture for the efficient and accurate prediction of water waves. The high-order spectral (HOS) method forms the foundation for the generation of the training data. The HOS method is applied for different, consecutive orders of nonlinearity starting from first order up to fourth order. The JONSWAP wave energy spectrum serves as the basis for modeling the one-dimensional irregular sea states. The overall objective of this work is to evaluate whether the complex non-linear physical processes can be identified and learned by the ML approach. The trained ML flow mapper is used to perform time integration of an initial sea state. The results indicate that the proposed ML approach is able to reproduce the distinctive physical processes of the different orders of nonlinearities. It is shown that the ML approach enables fast and accurate predictions of one-dimensional waves over a time horizon that spans multiple peak periods.

Machine-learning approach for discovery of conventional superconductors

First-principles computations are the driving force behind numerous discoveries of hydride-based superconductors, mostly at high pressures, during the last decade. Machine-learning (ML) approaches can further accelerate the future discoveries if their reliability can be improved. The main challenge of current ML approaches, typically aiming at predicting the critical temperature $T_{\rm c}$ of a solid from its chemical composition and target pressure, is that the correlations to be learned are deeply hidden, indirect, and uncertain. In this work, we showed that predicting superconductivity at any pressure from the atomic structure is sustainable and reliable. For a demonstration, we curated a diverse dataset of 584 atomic structures for which $\lambda$ and $\omega_{\log}$, two parameters of the electron-phonon interactions, were computed. We then trained some ML models to predict $\lambda$ and $\omega_{\log}$, from which $T_{\rm c}$ can be computed in a post-processing manner. The models were validated and used to identify two possible superconductors whose $T_{\rm c}\simeq 10-15$K at zero pressure. Interestingly, these materials have been synthesized and studied in some other contexts. In summary, the proposed ML approach enables a pathway to directly transfer what can be learned from the high-pressure atomic-level details that correlate with high-$T_{\rm c}$ superconductivity to zero pressure. Going forward, this strategy will be improved to better contribute to the discoveries of new superconductors.

AGRICO - Smart Way of Farming

Abstract: Research related to agriculture is growing rapidly, the challenges that lie ahead is solved with the help of advancement in technology. It is found that it is very beneficial for the economic growth and development of any nation. Especially in India, it requires the need for good research in order to improve agricultural productivity. In order to enhance the growth and solve the problems in the agricultural sector, the scientists use a variety of data mining methods. Different data mining techniques, like classification or prediction can be used to predict diseases in crops, and losses incurred as a result of these diseases. The diseases can be bacterial, fungal or viral. Some of the common plant diseases are bacterial wilt, black knot, curly top, etc. These diseases are caused by a variety of insects and micro- organisms. Our main focus in this research is on early identification of the diseases and helps the farmers in taking the decision to use the fertilizers that helps to protect the crops so that the diseases are eliminated in the early stage of production and so the farmers can get maximum yield. Ensemble method combines several classifiers to produce one finest predictive model and it is a very important technique in Machine Learning. In this paper, ensemble methods are used to predict the crop disease and an analyse has been done with the help of different classifiers such as Decision Trees, Naive Bayes Classifier, Random Forests, Support Vector Machine and K- Nearest Neighbour. Ensemble models, improves the performance of the classifiers that are weak. Te proposed machine learning approach that aims at predicting the best yielded crop for a particular region by analysing various atmospheric factors like rainfall, temperature, etc., and land factors like soil type including past records of crops grown. Finally, our system is expected to predict the best yielded crop based on dataset we have collected.

A supervised machine learning approach to discriminate the effect of carcass leachate on shallow groundwater quality around on-farm livestock mortality burial sites

The evaluation of leachate leakage at livestock mortality burial sites is challenging, particularly when groundwater is previously contaminated by agro-livestock farming. Supervised machine learning was applied to discriminate the impacts of carcass leachate from pervasive groundwater contamination in the following order: data labeling, feature selection, synthetic data generation, and classification. Physicochemical data of 359 water samples were collected from burial pits (LC), monitoring wells near pits (MW), pre-existing shallow household wells (HW), and background wells with pervasive contamination (BG). A linear classification model was built using two representative groups (LC and BG) affected by different pollution sources as labeled data. A classifier was then applied to assess the impact of leachate leakage in MW and HW. As a result, leachate impacts were observed in 40% of MW samples, which indicates improper construction and management of some burial pits. Leachate impacts were also detected in six HW samples, up to 120 m downgradient, within one year. The quantitative decision-making tool to diagnose groundwater contamination with leachate leakage can contribute to ensuring timely responses to leakage. The proposed machine learning approach can also be used to improve the environmental impact assessment of water pollution by improper disposal of organic waste.

Bipolar Disorder Detection using Machine Learning

Bipolar disorder, also known as manic-depressive illness, is a mental health condition that affects a person's mood, energy, and ability to function. People with bipolar disorder experience extreme shifts in their mood, ranging from periods of high energy and elation (called mania or hypomania) to periods of deep depression. This can interfere with a person's ability to work, study, or have healthy relationships. The diagnosis of bipolar disorder can be challenging due to the variability of symptoms and the lack of objective diagnostic tests. Machine learning algorithms have shown great potential in aiding the diagnosis of bipolar disorder by analyzing patterns in large datasets of clinical and neuroimaging data. In this paper, we present a machine learning approach for the detection of bipolar disorder using clinical and neuroimaging data. We applied feature selection and machine learning algorithms to classify patients with bipolar disorder from healthy controls. Our results showed that the proposed machine learning approach achieved an accuracy of 85% in classifying patients with bipolar disorder from healthy controls using clinical and neuroimaging data. These results suggest that machine learning can aid in the detection of bipolar disorder and may provide an objective diagnostic tool for clinicians.

Gastronomic image in the foodstagrammer’s eyes – A machine learning approach

Given the rich content that foodstagrammers, people who actively share their dining experiences using photographs and texts on social media, post, they considerably shape a destination's gastronomic image. Using big data analytics, this study examined the formation of gastronomic images from foodstagrammers' perspectives and the associated emotions. Moreover, it demonstrated the applicability of the proposed machine learning approach to evaluate both textual and pictorial content on social media. The study findings extend the current understanding of gastronomic images by identifying the underlying attributes based on the interplay of the three dimensions of food, environment, and activities. Furthermore, the results reveal specific image clusters and dimensions that arouse positive sentiments among foodstagrammers and influence users' engagement with the post. For practitioners, this study provides valuable insights into foodstagrammers' behaviors by identifying the aspects of gastronomic images that effectively arouse interest and engagement, thus promoting gastronomic destinations.

A machine learning approach for digital watermarking

ABSTRACT Recently, machine learning (ML) has been applying in almost all scientific fields to model and simulate the behaviour of complex systems. At the same time, the number of the proposed watermarking techniques have been increasing every year. Although most of the researchers in the watermarking and data hiding field put all their effort to compete each other to develop more efficient algorithm, the selection of the most efficient one is almost impossible for industries or software developers. In other words, comparison among all the watermarking techniques in order to select one watermarking technique would be cumbersome task. Moreover, the computer security is becoming more advanced which a single watermark technique cannot fulfil all the required criteria and there is a chance that an anti-watermarking is developed to remove the embedded watermark from the host data. This gap of knowledge to use the watermarking technology for a highly secure application is a real nightmare for the information security engineers and software designers. In this paper, two new approaches are proposed to train deep learning and shallow learning models based on the state-of-the-arts watermarking techniques. For these proposed approaches, several ML algorithms are applied to model various watermarked data, which are watermarked by different watermarking techniques in various spectrums and domains. An experimental setup is constructed based on several speeches, audios, images and videos beside the Amazon Sagemaker as ML modelling to implement the proposed approach. The experimental results show that apart from an overall effectiveness of the model, it would resolve some ambiguities by applying the proposed ML approach as a general watermarking workflow.

Full Waveform Inversion-Based Ultrasound Computed Tomography Acceleration Using Two-Dimensional Convolutional Neural Networks

AbstractUltrasound computed tomography (USCT) shows great promise in nondestructive evaluation and medical imaging due to its ability to quickly scan and collect data from a region of interest. However, existing approaches are a tradeoff between the accuracy of the prediction and the speed at which the data can be analyzed, and processing the collected data into a meaningful image requires both time and computational resources. We propose to develop convolutional neural networks (CNNs) to accelerate and enhance the inversion results to reveal underlying structures or abnormalities that may be located within the region of interest. For training, the ultrasonic signals were first processed using the full waveform inversion (FWI) technique for only a single iteration; the resulting image and the corresponding true model were used as the input and output, respectively. The proposed machine learning approach is based on implementing two-dimensional CNNs to find an approximate solution to the inverse problem of a partial differential equation-based model reconstruction. To alleviate the time-consuming and computationally intensive data generation process, a high-performance computing-based framework has been developed to generate the training data in parallel. At the inference stage, the acquired signals will be first processed by FWI for a single iteration; then the resulting image will be processed by a pre-trained CNN to instantaneously generate the final output image. The results showed that once trained, the CNNs can quickly generate the predicted wave speed distributions with significantly enhanced speed and accuracy.

A new method for predicting the prognosis of ischemic stroke based vascular structure features and lesion location features

ObjectiveDetermining the changes in the prognosis of the cerebral infarction area has an important guiding role in the selection of the treatment plan. The goal of this study is to propose a machine learning-based method that can predict the prognosis of stroke effectively and efficiently. Methods97 cases of stroke were analyzed retrospectively. Firstly, we extracted vascular structural features from computed tomography angiography (CTA) images and stroke location features from diffusion-weighted imaging (DWI) images to comprehensively characterize the lesions, respectively. Then, we performed sparse representation-based feature selection and classification to predict the prognosis of stroke based on the extracted features. Finally, we randomly divided the 97 cases into cross-validation set, independent testing set 1 and independent testing set 2 to validate the proposed model. Results464 vascular structure features and 116 positional features were extracted. After feature selection, 52 features were finally applied to build the classification model. The proposed model achieved promising prediction performance on the two independent testing sets, with the classification accuracies of 85.19% and 81.25%, respectively. ConclusionThe proposed machine learning approach can effectively mine and accurately quantify the features related to the prognosis, which include the vascular structural features and the stroke location features. In addition, the established prognostic prediction model based on these features has achieved interesting performances, which may provide valuable guidance for the clinical treatment of stroke.

Implicit Solutions of the Electrical Impedance Tomography Inverse Problem in the Continuous Domain with Deep Neural Networks.

Electrical impedance tomography (EIT) is a non-invasive imaging modality used for estimating the conductivity of an object Ω from boundary electrode measurements. In recent years, researchers achieved substantial progress in analytical and numerical methods for the EIT inverse problem. Despite the success, numerical instability is still a major hurdle due to many factors, including the discretization error of the problem. Furthermore, most algorithms with good performance are relatively time consuming and do not allow real-time applications. In our approach, the goal is to separate the unknown conductivity into two regions, namely the region of homogeneous background conductivity and the region of non-homogeneous conductivity. Therefore, we pose and solve the problem of shape reconstruction using machine learning. We propose a novel and simple jet intriguing neural network architecture capable of solving the EIT inverse problem. It addresses previous difficulties, including instability, and is easily adaptable to other ill-posed coefficient inverse problems. That is, the proposed model estimates the probability for a point of whether the conductivity belongs to the background region or to the non-homogeneous region on the continuous space Rd∩Ω with d∈{2,3}. The proposed model does not make assumptions about the forward model and allows for solving the inverse problem in real time. The proposed machine learning approach for shape reconstruction is also used to improve gradient-based methods for estimating the unknown conductivity. In this paper, we propose a piece-wise constant reconstruction method that is novel in the inverse problem setting but inspired by recent approaches from the 3D vision community. We also extend this method into a novel constrained reconstruction method. We present extensive numerical experiments to show the performance of the architecture and compare the proposed method with previous analytic algorithms, mainly the monotonicity-based shape reconstruction algorithm and iteratively regularized Gauss-Newton method.

A combinatorial machine-learning-driven approach for predicting glass transition temperature based on numerous molecular descriptors

ABSTRACT Glass transition temperature (Tg ) is one of the most significant thermophysical property which is hard to measure experimentally. With the development of machine learning, many molecular presentation methods and prediction algorithms have been proposed for predicting Tg . However, most descriptors of these algorithms are linear, while the values of molecular descriptors obtained from experiments or simulation software may have nonlinear relationships. General nonlinear machine learning methods are difficult to apply for the super high dimensional characteristics of molecular descriptors. In this paper, molecular descriptors are defined in manifold space and manifold learning is used for dimension reduction. Different from other linear algorithms for selecting molecular descriptors, Locally Linear Embedding (LLE) is utilised to reduce dimensionalities by extracting new features from numerous molecular descriptors. The features with dimensionality reduction are fed to the eXtreme Gradient Boosting (XGBoost) model. This ensemble algorithm is suitable for predicting glass transition temperature due to its high adaptability and nonlinear approximation ability. Genetic algorithm (GA) is adopted for optimising the parameters of XGBoost. The combinatorial approach consists of dimensionality reduction, prediction and optimisation. Experimental results show that the proposed machine learning approach LLE-XGBoost-GA has higher accuracy in Tg prediction of polymers.

Mobility Classification of LoRaWAN Nodes Using Machine Learning at Network Level.

LoRaWAN networks rely heavily on the adaptive data rate algorithm to achieve good link reliability and to support the required density of end devices. However, to be effective the adaptive data rate algorithm needs to be tuned according to the level of mobility of each end device. For that purpose, different adaptive data rate algorithms have been developed for the different levels of mobility of end devices, e.g., for static or mobile end devices. In this paper, we describe and evaluate a new and effective method for determining the level of mobility of end devices based on machine learning techniques and specifically on the support vector machine supervised learning method. The proposed method does not rely on the location capability of LoRaWAN networks; instead, it relies only on data always available at the LoRaWAN network server. Moreover, the performance of this method in a real LoRaWAN network is assessed; the results give clear evidence of the effectiveness and reliability of the proposed machine learning approach.

Optimized machine learning approaches for identifying vertical temperature gradient on ballastless track in natural environments

Due to the exposure of ballastless track to various geographical and meteorological conditions, the effects of temperature evolution on the performance of ballastless track should be considered. However, accurately predicting the vertical temperature gradient (VTG) on ballastless track in thermal environments has been challenging. This study develops four machine learning (ML) approaches, i.e., support vector regression (SVR), random forest (RF), extreme gradient boosting (XGBOOST), and artificial neural network (ANN), to identify the ballastless track’s VTG evolution in natural environments. The above ML approaches with hyperparameters optimized are trained and tested by 2000 samples from full-scale finite element simulation. The temperature field monitoring on ballastless track is performed to validate its temperature distribution simulation. The results show the XGBOOST-based method has the best accuracy for identifying the ballastless track’s VTG among the four selected ML approaches. Furthermore, the stacking-based hybrid framework is proposed to further improve the four selected ML approaches. For the stacking strategy, the improvement percentage of MAE of performance and transferability to the four selected ML approaches are 11.51%-55.43%, and 42.27%-77.79%, respectively. The proposed ML approaches are proven capable and efficient for predicting the temperature evolution of track engineering.