Search

Machine learning (ML) techniques are rapidly emerging as effective tools in predicting complex hydrological processes. The present study aims to comparatively assess the efficacy of four machine learning algorithms – Multi-Layer Perceptron (MLP), Extreme Gradient Boosting (XGBoost), Support Vector Regression (SVR), and Random Forest (RF) – in predicting water levels using rainfall data at the Batu Dam, Malaysia. Situated about 16 km from Kuala Lumpur city center, the Batu Dam plays a crucial role in flood mitigation and water supply. Utilizing a statistical approach, the models were evaluated based on key performance metrics: Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and the Coefficient of Determination (R2). Preliminary results accentuated the superior predictive prowess of the MLP model, especially for challenging forecasting scenarios with longer lag intervals. This investigation not only accentuates the potential of data-driven methodologies in hydrology but also offers valuable insights for water resource management in the region. When all scenarios for the MLP model are considered, it is observed that the 3-day scenario performed the best within MLP, with the lowest RMSE (at 0.0072) and MAE (at 0.005), and the highest R2 score (at 0.9972). Furthermore, within the MLP model. Due to its exceptionally high performance, the MLP-3 model proved to be an excellent choice for our modeling purposes. Furthermore, it was observed that MLP-3 yields a high R2 score of 0.994, and its predictions aligned closely with the actual water level values. This indicates that the model fits very well to the modeling problem. On the other hand, the SVR-30 model had an R2 score of 0.83, and its predictions were quite scattered with respect to the actual water levels. Four different input scenarios were investigated, considering correlation analysis. Generally, the comparison of four ML model indicated that the MLP model offered better accuracy in predicting daily water levels with respect to different assessment criteria. The findings of this study depicted the accomplishment of MLP model in capturing the changes in the water level of a dam thus paving the way for which the model can be used in works to mitigate potential risk that may occur in the future from natural events.

The classification of variable objects provides insight into a wide variety of astrophysics ranging from stellar interiors to galactic nuclei. The Zwicky Transient Facility (ZTF) provides time-series observations that record the variability of more than a billion sources. The scale of these data necessitates automated approaches to make a thorough analysis. Building on previous work, this paper reports the results of the ZTF Source Classification Project ( SCoPe ), which trains neural network and XGBoost (XGB) machine-learning (ML) algorithms to perform dichotomous classification of variable ZTF sources using a manually constructed training set containing 170,632 light curves. We find that several classifiers achieve high precision and recall scores, suggesting the reliability of their predictions for 209,991,147 light curves across 77 ZTF fields. We also identify the most important features for XGB classification and compare the performance of the two ML algorithms, finding a pattern of higher precision among XGB classifiers. The resulting classification catalog is available to the public, and the software developed for SCoPe is open source and adaptable to future time-domain surveys.

Accurate brain segmentation is critical for magnetic resonance imaging (MRI) analysis pipelines. Machine-learning-based brain MR image segmentation methods are among the state-of-the-art techniques for this task. Nevertheless, the segmentations produced by machine learning models often degrade in the presence of expected domain shifts between the test and train sets data distributions. These domain shifts are expected due to several factors, such as scanner hardware and software differences, technology updates, and differences in MRI acquisition parameters. Domain adaptation (DA) methods can make machine learning models more resilient to these domain shifts. This paper proposes a benchmark for investigating DA techniques for brain MR image segmentation using data collected across sites with scanners from different vendors (Philips, Siemens, and General Electric). Our work provides labeled data, publicly available source code for a set of baseline and DA models, and a benchmark for assessing different brain MR image segmentation techniques. We applied the proposed benchmark to evaluate two segmentation tasks: skull-stripping; and white-matter, gray-matter, and cerebrospinal fluid segmentation, but the benchmark can be extended to other brain structures. Our main findings during the development of this benchmark are that there is not a single DA technique that consistently outperforms others, and hyperparameter tuning and computational times for these methods still pose a challenge before broader adoption of these methods in the clinical practice.