Significance: Multilayer network clustering is used in such diverse areas as optimal islanding of critical infrastructures, analysis of trade agreements, and monitoring ecological interaction patterns. We propose a perspective on multilayer network clustering based on the concept of shape. By invoking the machinery of topological data analysis, we first study a shape of each node neighborhood and then group nodes based on how similar shapes of their local neighborhoods are. The significance of this methodology can be viewed through an emerging problem of sustainability of house insurance to climate risks. The topological perspective opens possibilities for more systematic, robust, and mathematically rigorous integration of higher-order network properties and their interplay to the analysis of complex networks.

The mosquito-borne dengue fever is a major public health problem in tropical countries, where it is strongly conditioned by climate factors such as temperature. In this paper, we formulate a holistic machine learning strategy to analyze the temporal dynamics of temperature and dengue data and use this knowledge to produce accurate predictions of dengue, based on temperature on an annual scale. The temporal dynamics are extracted from historical data by utilizing a novel multi-stage combination of auto-encoding, window-based data representation and trend-based temporal clustering. The prediction is performed with a trend association-based nearest neighbour predictor. The effectiveness of the proposed strategy is evaluated in a case study that comprises the number of dengue and dengue hemorrhagic fever cases collected over the period 1985–2010 in 32 federal states of Mexico. The empirical study proves the viability of the proposed strategy and confirms that it outperforms various state-of-the-art competitor methods formulated both in regression and in time series forecasting analysis.

An Argo‐based estimate of oxygen (O₂) at 150 m is presented for the Southern Ocean (SO) from temperature (T), salinity (S), and O₂ Argo profiles collected during 2008–2012. The method is based on a supervised machine learning algorithm known as random forest (RF) regression and provides an estimate for O₂ given T, S, location, and time information. The method is validated by attempting to reproduce the Southern Ocean State Estimate (SOSE) O₂ field using synthetic data sampled from SOSE. The RF mapping shows skill in the majority of the domain but is problematic in some of the boundary regions. Maps of O₂ at 150 m derived from observed profiles suggest that SOSE and the World Ocean Atlas 2013 climatology may overestimate annual mean O₂ in the SO, both on a global and basin scale. A large regional bias is found east of Argentina, where high O₂ values in the Argo‐based estimate are confined closer to the coast compared to other products. SOSE may also underestimate the annual cycle of O₂. Evaluation of the RF‐based method demonstrates its potential to improve understanding of O₂ annual mean fields and variability from sparse O₂ measurements. This implies the algorithm will also be effective for mapping other biogeochemical variables (e.g., nutrients and carbon). Furthermore, our RF evaluation results can be used to inform the design of future enhancements to the current array of O₂ profiling floats.