Multi-Model Adaptation Learning With Possibilistic Clustering Assumption for EEG-Based Emotion Recognition.

Extreme weather modeling is a challenge for modeling experts in Indonesia and the world. Extreme weather prediction is a complex problem because the chances of it happening are very small, so the developed models often have a low level of accuracy. The purpose of this research is to combine the classic model, Autoregressive Integrated Moving Average (ARIMA), recurrent neural network (RNN) model using Adam and SGD estimation (RNN-Adam and RNN-SGD) with the reLU, tanh, sigmoid and gaussian activation functions. In addition, the ARIMA-RNN mix model was also demonstrated in this study. These models are applied to monthly period extreme weather data obtained from the Meteorology, Climatology and Geophysics Agency (BMKG) of West Sulawesi Province which are converted into training data and test data. The RMSE value is used to see the level of prediction accuracy in both training data and test data. Based on the research results, the best model obtained for modeling Indonesia’s extreme weather is the ARIMA-RNN-Adam mix model with the reLU activation function based on the RMSE value on the training and test data. At n = 50, the smallest RMSE and MSE values of the third model are the ARIMA-RNN-Adam model which is 0.23212 using the reLU activation function, then the ARIMA-RNN-SGD model which is 0.25432 with the same activation function, while the ARIMA value is 0.3270. At n=100 it can be seen that the smallest RMSE and MSE values of the three models are the ARIMA-RNN-Adam model which is equal to 0.25149 using the reLU activation function, then the ARIMA-RNN-SGD model which is equal to 0.25256 with the same activation function, while the ARIMA value is 0.2644.

The effectiveness of conventional deep learning-based intelligent fault diagnosis models depends on the training data and testing data following the same probability distribution. But the discrepancy in cross-domain distributions is inherent because of changes in external and internal conditions, resulting in a decline in diagnosis performance. Recently, transfer learning is employed to induce an adaptive diagnosis network in the scenario of distribution discrepancies. However, little attention has been paid to fully consider the cross-layer interaction and feature transferability for traditional transfer learning-based diagnosis networks. To overcome these problems, this paper presents a novel transferable bilinear neural network for cross-domain diagnosis. First, the bilinear map between bi-layer features is used to implement a novel information fusion and significantly improves the feature representation capability. It also realizes the embedding of bi-layer joint distributions into the reproducing kernel Hilbert space. Based on the embedding and feature transferability analysis, a reliable adaptive framework is designed to enable effective cross-domain transfer learning. The effectiveness of the proposed approach is validated using experiments with various transfer scenarios.

Financial institutions have a large amount of data on their borrowers, which can be used to predict the probability of borrowers defaulting their loan or not. Some of the models that have been used to predict individual loan defaults include linear discriminant analysis models and extreme value theory models. These models are parametric in nature since they assume that the response being investigated takes a particular functional form. However, there is a possibility that the functional form used to estimate the response is very different from the actual functional form of the response. The purpose of this research was to analyze individual loan defaults in Kenya using the logistic regression model. The data used in this study was obtained from equity bank of Kenya for the period between 2006 to 2016. A random sample of 1000 loan applicants whose loans had been approved by equity bank of Kenya during this period was obtained. Data obtained was on the credit history, purpose of the loan, loan amount, nature of the saving account, employment status, sex of the applicant, age of the applicant, security used when acquiring the loan and the area of residence of the applicant (rural or urban). This study employed a quantitative research design, it deals with individual loans defaults as group characteristics of a borrower. The data was pre-processed by seeding using R- Software and then split into training dataset and test data set. The train data was used to train the logistic regression model by employing Supervised machine learning approach. The R-statistical software was used for the analysis of the data. The test data set was used to do cross-validation of the developed logistic model which later was used for analysis prediction of individual loan defaults. This study focused on the analysis of individual loan defaults in Kenya using the logistic regression model in Machine learning. The logistic regression model predicted 303 defaults from train data set, 122 non-defaults and misclassified loans were 56 and 69. The model had an accuracy of 0.7727 with the train data and 0.7333 with the test data. The logistic regression model showed a precision of 0.8440 and 0.8244 with the train and test data respectively. The performance of the model with both the train and test data was illustrated using a plot of train errors and test errors against sample size on the same axes. The plot showed that the performance of the model increases with an increase in sample size. The study recommended the use of logistic regression in conjunction with supervised machine learning approach in loan default prediction in financial institutions and also more research should be carried out on ensemble methods of loan defaults prediction in order to increase the prediction accuracy.

Automated monitoring of the movements and behaviour of animals is a valuable research tool. Recently, machine learning tools were applied to many species to classify units of behaviour. For the monitoring of wild species, collecting enough data for training models might be problematic, thus we examine how machine learning models trained on one species can be applied to another closely related species with similar behavioural conformation. We contrast two ways to calculate accuracies, termed here as overall and threshold accuracy, because the field has yet to define solid standards for reporting and measuring classification performances. We measure 21 dogs and 7 wolves, and find that overall accuracies are between 51 and 60% for classifying 8 behaviours (lay, sit, stand, walk, trot, run, eat, drink) when training and testing data are from the same species and between 41 and 51% when training and testing is cross-species. We show that using data from dogs to predict the behaviour of wolves is feasible. We also show that optimising the model for overall accuracy leads to similar overall and threshold accuracies, while optimizing for threshold accuracy leads to threshold accuracies well above 80%, but yielding very low overall accuracies, often below the chance level. Moreover, we show that the most common method for dividing the data between training and testing data (random selection of test data) overestimates the accuracy of models when applied to data of new specimens. Consequently, we argue that for the most common goals of animal behaviour recognition overall accuracy should be the preferred metric. Considering, that often the goal is to collect movement data without other methods of observation, we argue that training data and testing data should be divided by individual and not randomly.

Hypertensive disorders of pregnancy (HDP) cause adverse maternal and fetal outcomes, and establishing an early prediction method for HDP is needed. Existing methods based on serum markers require blood sampling. Therefore, we aimed to develop and validate a noninvasive HDP prediction model based on home blood pressure monitoring. In a development cohort, home blood pressure monitoring data from 443 pregnant women including 65 HDPs were divided into training data (n=365) and test data (n=78) to develop a logistic regression-based prediction model. Normal blood pressure variations depending on season and gestational age were subtracted. At each time point, 4 features were calculated for the last 4 weeks of data: average systolic blood pressure, average diastolic blood pressure, correlation coefficient between systolic blood pressure and diastolic blood pressure, and upward trend of systolic blood pressure against day. In a validation cohort, HBPM data from 264 pregnant women including 33 HDPs were collected prospectively and used to validate the model. The area under the receiver operating characteristic curve was 0.949 (95% CI, 0.890-1.000), 0.884 (95% CI, 0.807-0.961), and 0.845 (95% CI, 0.759-0.930) for the training, test, and validation data, respectively. Sensitivity, specificity, positive predictive value, and negative predictive value were 0.923, 0.888, 0.387, and 0.993 for the training data; 0.667, 0.897, 0.867, and 0.729 for the test data; 0.758, 0.766, 0.316, and 0.957 for the validation data. Our HDP prediction model based on home blood pressure monitoring showed high negative predictive values and may contribute to reducing medical consultations.

Passive sensing applications are limited by their inability to determine who is using a device, a critical concern in child mobile device use research, where devices are often shared between siblings or between a child and their parent. Our previous work leveraged behavioral biometrics to identify a target child user; however, it is unknown what type of training data is necessary for optimal model performance. This study evaluated model performance across different characteristics of training data. Thirty-six children (11.3 ± 0.9 years, 56% female) self-selected a video or a game on iPads for 10 min while laying and for another 5 min while sitting. The SensorLog application captured iPad accelerometer and gyroscope data while the child interacted with the device. Machine learning algorithms including Neural Network (NN), Random Forest (RF), k-Nearest Neighbors (k-NN), and SwipeFormer were applied to determine the most important aspects of training data to optimize model performance. The aspects of training data evaluated included (1) varying the length (i.e., seconds of training data), (2) varying the user position (i.e., sitting, laying), and (3) varying the time proximity between training and testing data. F1 score was used to evaluate model performance. The SwipeFormer F1 scores were lowest when the training data was further from the test data (0 when training data was 11 min away from test data) and highest when training data was close to test data (0.91 when training data was the minute preceding test data). The SwipeFormer F1 scores were highest when predicting the user laying from laying (0.97) and sitting from sitting (0.94), and lowest when predicting the user sitting from laying (0) and laying from sitting (0). The length of training data had little impact on performance, with a SwipeFormer F1 score of 0.91 when training on one minute of data and a SwipeFormer F1 score of 0.94 when training on twelve minutes of data. Because researchers would likely be predicting users at different timepoints than their training data, research should focus on improving model performance for identifying users independent of time proximity for training and test data.

Dimensionality reduction has emerged as one of the prominent fields of research since it provides a solution to a wide class of problems, such as compression, classification, regression, feature analysis, and visual recognition. Generally, subspace learning algorithms find a low-dimensional subspace from a given high-dimensional data, wherein samples from different classes can be well separated. In the past decades, several types of dimensionality reduction and subspace selection algorithms have been widely used for visual data analysis. Conventional algorithms such as principal component analysis (PCA) and linear discriminant analysis (LDA), perform well under the assumption that training and test data follow similar distributions. However, the main drawback still remains in the sense that the distribution of the training data and that of the test data are mismatched when samples are drawn from different but related sources. They also fail to incorporate temporal statistical variations in data distribution when applied to time-series data(videos). Consequently they result in unsatisfactory recognition performance when they deal with real world visual analysis problems. To overcome the above issues, in this research, our aim is to consider invariance and stationarity in subspace analysis pertaining to many computer vision applications such as video retrieval, human behaviour analysis, event analysis and activity recognition. More specifically, we will address the challenging and fundamental tasks of 1. Video Classification (Scene, dynamic texture and action recognition) 2. Visual Domain Adaptation. To address the first task, we propose a subspace learning approach that focuses on extracting the stationary parts of all videos in the same class. The notion of stationarity is intuitively well-adapted to model the temporal nature of the video signal and lets us make use of many image features. Instead of modeling temporal information in the features, our method explicitly accounts for it when learning the latent space. As a consequence, the resulting video representation is particularly well-suited for classification purpose. Our experimental evaluation shows that our approach outperforms baselines from the different groups of methods in several video classification tasks such as dynamic texture recognition, scene classification and action recognition.The second task addressed by this work is domain adaptation for visual recognition. In our proposed approach, we follow the intuitive idea of learning the domain invariant subspace by matching the distributions of the transformed training and test data using Maximum Mean Discrepancy (MMD). This, we believe, makes better use of the expressive power of the kernel in MMD compared to the other approaches like Transfer Component Analysis (TCA). Although motivated by MMD, in TCA, the distance between the sample means is measured in a lower-dimensional space rather than in Reproducing Kernel Hilbert Space (RKHS), which somewhat contradicts the intuition behind the use of kernels. Furthermore, we extend the framework for the semi-supervised scenario. Experiments on benchmark domain adaptation dataset for visual recognition, show that in comparison to well-known methods, the proposed approach obtains a significant improvement in classification accuracy. To address the second task, we propose another domain adaptation method that exploits the Riemannian structure of the statistical manifold when learning the invariant subspaces or samples. To this end, we introduce the use of the Hellinger distance which is related to the geodesic distance on the space of probability distributions. While the Hellinger distance has been employed for dimensionality reduction, to the best of our knowledge, our approach is the first attempt at exploiting the Riemannian geometry of the statistical manifold for domain adaptation. We showed that our sample-selection and subspace-based approaches in conjunction with Hellinger distance for distribution matching consistently outperform the similar approaches on the tasks of visual domain adaptation and WiFi localization.

Quantification and solutions of arm movements effect on sEMG pattern recognition

The Sunda Strait is an important area for Indonesia because it is the main domestic and international transportation route. As a water area, the Sunda Strait has weather conditions that are greatly influenced by sea surface temperature (SST). Crucial SST forecasting is carried out to assist maritime transportation activities. This study aims to compare the performance of the ARIMA and LSTM methods in forecasting SST in the Sunda Strait. The data used in this study are daily SST data for the Sunda Strait from August 20, 2022, to January 1, 2024. The best ARIMA model obtained in this data modeling is ARIMA(1,1,1), where this model has significant overall parameters, the smallest AIC and BIC values, and model diagnostic results that meet the assumptions. Meanwhile, in LSTM modeling, the best combination of hyperparameters obtained is a neuron of 150, an epoch of 150, and a batch size of 32, where this combination produces the lowest MSE value of 0.003799. A comparison of the performance of the ARIMA and LSTM methods is carried out by considering the MAPE values ​​of the training data and test data. The LSTM method is superior to the ARIMA method, with a MAPE value of 0.512% for training data and 0.564 for testing data. The forecasting results using the LSTM method show a similar pattern in both training and testing data. Meanwhile, the forecasting results of the LSTM method for the following 30 periods show a fluctuating pattern throughout the day.

Currently, much research on machine learning about prediction has been carried out. For example, to predict the exchange rate of the rupiah against the United States currency, namely the United States Dollar (USD). The continuing trend of USD depreciation has attracted many researchers to explore currency trading, especially in establishing an efficient method for predicting fluctuating exchange rates. The rapid development of time series prediction methods has resulted in many methods that can predict data according to needs. In this study, we apply the Fuzzy Time Series Cheng method with Particle Swarm Optimization (PSO) to predict the IDR exchange rate against USD. The data used in this research is sourced from Bank Indonesia in the form of time series data on the selling and buying exchange rate. The FTS Cheng method forecasts the IDR exchange rate against USD. In contrast, the PSO algorithm optimizes the interval parameter to increase the forecasting accuracy. Based on the implementation and the results of the tests, the results show that using the PSO algorithm can produce the best optimization interval parameters and increase the accuracy value. From the results of 10 trials with training data, testing data, and different iterations, it was obtained that the MAPE test for predicting the rupiah exchange rate against the US dollar using FTS Cheng with 60% training data and 40% testing data resulted in the lowest MAPE of 0.610145%. Furthermore, 70% of the training and 30% of the testing data resulted in the lowest MAPE of 0.313388%. Then the FTS Cheng and PSO testing with 60% training data and 40% testing data, and an iteration value of 200 resulted in the lowest MAPE of 0.394707%. Furthermore, 70% of training data and 30% of testing data and an iteration value of 90 resulted in the lowest MAPE of 0.263666%.

In recent years, emotion recognition, particularly EEG-based emotion recognition, has found widespread application across various domains. Enhancing EEG data processing and emotion recognition models remains a key research focus in this field. This paper presents an emotion recognition framework combining the CUSUM algorithm-based adaptive window selection technique with the convolutional attention-enhanced Kolmogorov-Arnold Networks (CA-KAN). The improved CUSUM algorithm effectively extracts the most emotion-relevant segments from raw EEG data. Furthermore, by enhancing the KAN network, the CA-KAN model achieves both high accuracy and efficiency in emotion recognition. The proposed framework achieved peak classification accuracies of 94.63% and 94.73% on the SEED and SEED-IV datasets, respectively. Additionally, the framework offers a lightweight advantage, demonstrating significant potential for real-world applications, including medical emotion monitoring and driver emotion detection.

Cross-domain vulnerability detection using graph embedding and domain adaptation

Data-driven approach for hydrocarbon production forecasting using machine learning techniques

Building materials is an important factor to built a house, to estimate funds the needs of build a house, consumers or developers can estimate the funds needed to build a house. To solve these problems use case base reasoning (CBR) approach, which method is capable of reasoning or solving the problem based on the cases that have been there as a solution to new problems. The system built in this study is a CBR system for determine the needs of house building materials. The consultation process is done by inserting new cases compared to the old case similarity value is then calculated using the nearest neighbor.
 The first test by inserting test data then compared with each type of home then obtained an accuracy of 83.6%. The second test is done by K-fold Cross Validation with K = 25 with the number of data 200, the data will be divided into two parts, namely the training data and test data, training data as many as 192 data and test data as many as 8 data. K-Fold Cross Validation method. This CBR system can produce an accuracy of 85.71%

As a maritime country strategically located along the world's leading transportation routes, Indonesia often faces increased ship accidents. Based on the Basarnas Statistics Book, ship accidents handled by Basarnas from 2021 to 2023 increased by 3%. This condition requires an effective forecasting method to carry out SAR operations to predict ship accidents in the Indonesian region in the future and assess the readiness and needs of Basarnas resources. This study compares the forecasting results obtained using the Singular Spectrum Analysis (SSA) and the Local Linear methods. Both methods do not require parametric assumptions. The data used in this study are divided into training data and test data. This data is secondary data obtained from the Basarnas Statistics Book. The training data in this study is the number of SAR operations from January 2021 to December 2022, while the testing data is from January 2023 to December 2023. From the analysis results, it is known that the method with the smallest MAPE is the Local Linear method with a MAPE of test data of 18.67% (good forecasting category), optimal bandwidth (h) = 4.299, and CV (h) = 231.39 where bandwidth is used to determine the level of smoothness of the estimate, while the CV (h) value is used to select the optimal bandwidth that minimizes the estimation error. At the same time, the SSA method has a MAPE of 40.27% (fair forecasting category). This shows that the Local Linear method provides a more accurate forecast of the number of SAR operations related to ship accidents in Indonesia. This research contributes to the SDGs to make Basarnas an effective and accountable institution and improve the planning and decision-making process in SAR operations through accurate forecasting research is relevant to accurate forecasting.