Sparse Data Research Articles

SEBGM: Sentence Embedding Based on Generation Model with multi-task learning

Sentence embedding, which aims to learn an effective representation of a sentence, is a significant part for downstream tasks. Recently, using contrastive learning and pre-trained model, most methods of sentence embedding achieve encouraging results. However, on the one hand, these methods utilize discrete data augmentation to obtain positive samples performing contrastive learning, which could distort the original semantic of sentences. On the other hand, most methods directly employ the contrastive frameworks of computer vision to perform contrastive learning, which could confine the contrastive training due to the discrete and sparse text data compared with image data. To solve the issues above, we design a novel contrastive framework based on generation model with multi-task learning by supervised contrastive training on the dataset of natural language inference (NLI) to obtain meaningful sentence embedding (SEBGM). SEBGM makes use of multi-task learning to enhance the usage of word-level and sentence-level semantic information of samples. In this way, the positive samples of SEBGM are from NLI rather than data augmentation. Extensive experiments show that our proposed SEBGM can advance the state-of-the-art sentence embedding on the semantic textual similarity (STS) tasks by utilizing multi-task learning.

Heterogeneous Adaptive Preference Learning for Recommendation

Graph-based collaborative filtering techniques have emerged as a promising recommendation approach by modeling user-item interaction as graphs. Recently, contrastive learning has been employed in graph collaborative through data augmentation, which can effectively offer data efficiency and reduce labeling costs. Nonetheless, most existing contrastive learning approaches overlook the heterogeneous auxiliary information pertaining to users and items, such as user social relationships and item categories, which are crucial to alleviate the data sparsity issue. In this paper, we propose a novel contrastive learning method, referred to as Heterogeneous Adaptive Preference Learning for Recommendation (HAPLRec), which explicitly incorporates fine-grained preference information from both users and items. Specifically, we construct user relationship graphs and item relationship graphs based on specific meta-paths in a heterogeneous graph. Subsequently, we conduct data augmentation on these graphs individually to obtain auxiliary contrastive tasks. Moreover, we introduce an optimization algorithm that leverages the gradient similarity between the main task and the auxiliary tasks, dynamically adjusting the weight assigned to each task to expedite achieving superior performance within a shorter time frame. The effectiveness of the proposed model is demonstrated through extensive experiments conducted on three publicly available datasets.

Mild cognitive impairment in Parkinson's disease: current view

Parkinson's disease (PD), the most common motor movement disorder and second most common neurodegenerative disorder after Alzheimer's disease (AD), is often preceded by a period of mild cognitive impairment (MCI), which is associated with impairment of a variety of cognitive domains including executive function, attention, visuospatial abilities and memory. MCI, a risk factor for developing dementia, affects around 30% of de novo PD patients and can increase to 75% after more than 10 years. While 30–40% remain in the MCI state, up to 60% will convert to dementia. Characteristic findings are slowing of EEG rhythms, frontotemporal hypoperfusion, decreased functional connectivity in the default mode and attentional networks, prefrontal and basal-ganglia-cortical circuits, which often manifests prior to clinical symptoms and overt brain atrophy. The heterogeneity of cognitive phenotypes suggests that a common neurodegenerative process affects multiple functional neuronal networks and neuromodulatory systems that may be superimposed by Lewy body and Alzheimer's-related or other co-pathologies. Sparse neuropathological data for PD-MCI revealed a heterogenous picture with various morphological changes similar to MCI in other diseases. This review highlights the essential epidemiological, clinical, neuroimaging and morphological changes in PD-MCI, available biomarkers, and discusses the heterogenous pathobiological mechanisms involved in its development. In view of its complex pathogenesis, well-designed longitudinal clinico-pathological studies are warranted to clarify the alterations leading to MCI in PD, which may be supported by fluid and neuroimaging biomarkers as a basis for early diagnosis and future adequate treatment modalities of this debilitating disorder.

Utilizing domain knowledge: Robust machine learning for building energy performance prediction with small, inconsistent datasets

Machine learning (ML) applications often require large datasets, a requirement that can pose a major challenge in fields where data is sparse or inconsistent. To address this issue, we propose a novel approach that combines prior knowledge with data-driven methods to significantly reduce data dependency. This study represents a disentangled system compositionality knowledge by the method of Component-Based Machine Learning (CBML) in the context of energy-efficient building engineering. In this way, CBML incorporates semantic domain knowledge within the structure of a data-driven model. To understand the advantage of CBML, we conducted a case experiment to assess the effectiveness of this knowledge-encoded ML approach in scenarios with sparse data input (1 % - 0.0125 % sampling rate) and several typical ML methods. Our findings reveal three key advantages of this approach over traditional ML methods: 1) It significantly improves the robustness of ML models when dealing with extremely small and inconsistent datasets; 2) It allows for efficient utilization of data from diverse record collections; 3) It can handle incomplete data while maintaining high interpretability and reducing training time. These features offer a promising solution to the challenges associated with deploying data-intensive methods and contribute to more efficient real-world data usage. Additionally, we outline four essential prerequisites to ensure the successful integration of prior knowledge and ML generalization in target scenarios and open-sourced the code and dataset for community reproduction.

A two-step item bank calibration strategy based on 1-bit matrix completion for small-scale computerized adaptive testing.

Computerized adaptive testing (CAT) is a widely embraced approach for delivering personalized educational assessments, tailoring each test to the real-time performance of individual examinees. Despite its potential advantages, CAT�s application in small-scale assessments has been limited due to the complexities associated with calibrating the item bank using sparse response data and small sample sizes. This study addresses these challenges by developing a two-step item bank calibration strategy that leverages the 1-bit matrix completion method in conjunction with two distinct incomplete pretesting designs. We introduce two novel 1-bit matrix completion-based imputation methods specifically designed to tackle the issues associated with item calibration in the presence of sparse response data and limited sample sizes. To demonstrate the effectiveness of these approaches, we conduct a comparative assessment against several established item parameter estimation methods capable of handling missing data. This evaluation is carried out through two sets of simulation studies, each featuring different pretesting designs, item bank structures, and sample sizes. Furthermore, we illustrate the practical application of the methods investigated, using empirical data collected from small-scale assessments.

Laser-irradiating infrared attenuated total reflection spectroscopy of articular cartilage: Potential and challenges for diagnosing osteoarthritis

ObjectiveA prototype infrared attenuated total reflection (IR-ATR) laser spectroscopic system designed for in vivo classification of human cartilage tissue according to its histological health status during arthroscopic surgery is presented. Prior to real-world in vivo applications, this so-called osteoarthritis (OA) scanner has been tested at in vitro conditions revealing the challenges associated with complex sample matrices and the accordingly obtained sparse spectral datasets. MethodsIn vitro studies on human knee cartilage samples at different contact pressures (i.e., 0.2–0.5 ​MPa) allowed recording cartilage degeneration characteristic IR signatures comparable to in vivo conditions with high temporal resolution. Afterwards, the cartilage samples were assessed based on the clinically acknowledged osteoarthritis cartilage histopathology assessment (OARSI) system and correlated with the obtained sparse IR data. ResultsAmide and carbohydrate signal behavior was observed to be almost identical between the obtained sparse IR data and previously measured FTIR data used for sparse partial least squares discriminant analysis (SPLSDA) to identify the spectral regions relevant to cartilage condition. Contact pressures between 0.3 and 0.4 ​MPa seem to provide the best sparse IR spectra for cylindrical (d ​= ​3 ​mm) probe tips. ConclusionLaser-irradiating IR-ATR spectroscopy is a promising analytical technique for future arthroscopic applications to differentiate healthy and osteoarthritic cartilage tissue. However, this study also revealed that the flexible connection between the laser-based analyzer and the arthroscopic ATR-probe via IR-transparent fiberoptic cables may affect the robustness of the obtained IR data and requires further improvements.

A Data Assimilation Methodology to Analyze the Unsaturated Seepage of an Earth–Rockfill Dam Using Physics-Informed Neural Networks Based on Hybrid Constraints

Data assimilation for unconfined seepage analysis has faced significant challenges due to hybrid causes, such as sparse measurements, heterogeneity of porous media, and computationally expensive forward models. To address these bottlenecks, this paper introduces a physics-informed neural network (PINN) model to resolve the data assimilation problem for seepage analysis of unsaturated earth–rockfill dams. This strategy offers a solution that decreases the reliance on numerical models and enables an accurate and efficient prediction of seepage parameters for complex models in the case of sparse observational data. For the first attempt in this study, the observed values are obtained by random sampling of numerical solutions, which are then contributed to the synchronous constraints in the loss function by informing both the seepage control equations and boundary conditions. To minimize the effects of sharp gradient shifts in seepage parameters within the research domain, a residual adaptive refinement (RAR) constraint is introduced to strategically allocate training points around positions with significant residuals in partial differential equations (PDEs), which could facilitate enhancing the prediction accuracy. The model’s effectiveness and precision are evaluated by analyzing the proposed strategy against the numerical solutions. The results indicate that even with limited sparse data, the PINN model has great potential to predict seepage data and identify complex structures and anomalies inside the dam. By incorporating coupling constraints, the validity of our PINN model could lead to theoretically viable applications of hydrogeophysical inversion or multi-parameter seepage inversion. The results show that the proposed framework can predict the seepage parameters for the entire research domain with only a small amount of observation data. Furthermore, with a small amount of observation data, PINNs are able to obtain more accurate results than purely data-driven DNNs.

A parallel and performance portable implementation of a full-field crystal plasticity model

We have developed a parallel implementation of an Elasto-Viscoplastic Fast Fourier Transform-based (EVPFFT) micromechanical solver to enable computationally efficient crystal plasticity modeling for polycrystalline materials. Our primary focus lies in achieving performance portability, allowing a single EVPFFT implementation to run optimally on various homogeneous architectures, including multi-core Central Processing Units (CPUs), as well as on heterogeneous computer architectures comprising multi-core CPUs and Graphics Processing Units (GPUs) from different vendors. To accomplish this goal, we have leveraged MATAR, a C++ software library that simplifies the creation and utilization of multidimensional dense or sparse matrix and array data structures. These data structures are designed to be portable across diverse architectures through the use of Kokkos, a performance-portable library. Additionally, we have employed the Message Passing Interface (MPI) to efficiently distribute the computational workload among processors. The heFFTe (Highly Efficient FFT for Exascale) library is used to facilitate the performance portability of the fast Fourier transforms (FFTs) computation. The computational performance of EVPFFT is evaluated and presented in terms of parallel scalability and simulation runtime on different high-performance computing (HPC) architectures. The utility of the developed framework to efficiently simulate the micro-mechanical fields in polycrystalline microstructures in engineering applications is discussed. Program summaryProgram Title: EVPFFTCPC Library link to program files:https://doi.org/10.17632/2k8579fyyv.1Developer's repository link:https://github.com/lanl/FierroLicensing provisions: BSD 3-Clause LicenseProgramming language: C++External routines/libraries: MPI, Kokkos, MATAR, HeFFTe, HDF5Nature of problem: EVPFFT is a crystal plasticity code designed to compute micro-mechanical fields within a polycrystalline representative volume element (RVE) and predict the macroscale response of the RVE.Solution method: EVPFFT uses the periodic Green's function method in Fourier space to solve the field equations of static stress equilibrium in a periodic spatial domain.

A semi-supervised approach of short text topic modeling using embedded fuzzy clustering for Twitter hashtag recommendation

Social media stands as a crucial information source across various real-world challenges. Platforms like Twitter, extensively used by news outlets for real-time updates, categorize news via hashtags. These hashtags act as pivotal meta-information for linking tweets to underlying themes, yet many tweets lack them, posing challenges in topic searches. Our contribution addresses this by introducing a novel heuristic for hashtag recommendation. Extracting 20 thousand tweets, 5000 each from distinct categories health, sports, politics, and technology we applied fundamental data cleaning and tokenization techniques. Leveraging Word2Vec, we vectorized tokens, capturing nuanced semantic meanings and mitigating data sparsity issues. The proposed heuristic creates clusters of different topic by combining these embedded features and idea of fuzzy C-Means technique. Develop a rule-based approach that combines both supervised and unsupervised methods to label clusters, indicating their respective topic. The experimental outcomes shows that our proposed techniques achieve better performance metrics in precision, recall, and F1-score compared to specific baseline models.

Cross-domain information fusion and personalized recommendation in artificial intelligence recommendation system based on mathematical matrix decomposition

Given the challenges of inter-domain information fusion and data sparsity in collaborative filtering algorithms, this paper proposes a cross-domain information fusion matrix decomposition algorithm to enhance the accuracy of personalized recommendations in artificial intelligence recommendation systems. The study begins by collecting Douban movie rating data and social network information. To ensure data integrity, Levenshtein distance detection is employed to remove duplicate scores, while natural language processing technology is utilized to extract keywords and topic information from social texts. Additionally, graph convolutional networks are utilized to convert user relationships into feature vectors, and a unique thermal coding method is used to convert discrete user and movie information into binary matrices. To prevent overfitting, the Ridge regularization method is introduced to gradually optimize potential feature vectors. Weighted average and feature connection techniques are then applied to integrate features from different fields. Moreover, the paper combines the item-based collaborative filtering algorithm with merged user characteristics to generate personalized recommendation lists.In the experimental stage, the paper conducts cross-domain information fusion optimization on four mainstream mathematical matrix decomposition algorithms: alternating least squares method, non-negative matrix decomposition, singular value decomposition, and latent factor model (LFM). It compares these algorithms with the non-fused approach. The results indicate a significant improvement in score accuracy, with mean absolute error and root mean squared error reduced by 12.8% and 13.2% respectively across the four algorithms. Additionally, when k = 10, the average F1 score reaches 0.97, and the ranking accuracy coverage of the LFM algorithm increases by 54.2%. Overall, the mathematical matrix decomposition algorithm combined with cross-domain information fusion demonstrates clear advantages in accuracy, prediction performance, recommendation diversity, and ranking quality, and improves the accuracy and diversity of the recommendation system. By effectively addressing collaborative filtering challenges through the integration of diverse techniques, it significantly surpasses traditional models in recommendation accuracy and variety.

Tracking long-term modal behaviour of a footbridge and identifying potential SHM approaches

AbstractNumerous studies have investigated the long-term monitoring of natural frequencies, primarily focusing on medium–large highway bridges, using expensive monitoring systems with a large array of sensors. However, this paper addresses the less explored issue of monitoring a footbridge, examining four critical aspects: (i) sensing system, (ii) frequency extraction method, (iii) data modelling techniques, and (iv) damage detection. The paper proposes a low-cost all-in-one sensor/logger unit instead of a conventional sensing system to address the first issue. For the second issue, many studies use natural frequency data extracted from measured acceleration for data modelling, the paper highlights the impact of the input parameters used in the automated frequency extraction process, which affects the number and quality of frequency data points extracted and subsequently influences the data models that can be created. After that, the paper proposes a modified PCA model optimised for computational efficiency, designed explicitly for sparse data from a low-cost monitoring system, and suitable for future on-board computation. It also explores the capabilities and limitations of a data model developed using a limited data set. The paper demonstrates these aspects using data collected from a 108 m cable-stayed footbridge over several months. Finally, the detection of damage is achieved by employing the one-class SVM machine learning technique, which utilises the outcomes obtained from data modelling. In summary, this paper addresses the challenges associated with the long-term monitoring of a footbridge, including selecting a suitable sensing system, automated frequency extraction, data modelling techniques, and damage detection. The proposed solutions offer a cost-effective and efficient approach to monitoring footbridges while considering the challenges of sparse data sets.

Collaborative graph contrastive learning for recommendation

Recently, Graph Neural Networks (GNNs) using aggregating neighborhood collaborative information have shown effectiveness in recommendation. However, GNNs-based models suffer from over-smoothing and data sparsity problems. Due to its self-supervised nature, contrastive learning has gained considerable attention in the field of recommendation, aiming at alleviating highly sparse data. Graph contrastive learning models are widely used to learn the consistency of representations by constructing different graph augmentation views. Most current graph augmentation with random perturbation destroy the original graph structure information, which mislead embeddings learning. In this paper, an effective graph contrastive learning paradigm CollaGCL is proposed, which constructs graph augmentation by using singular value decomposition to preserve crucial structure information. CollaGCL enables perturbed views to effectively capture global collaborative information, mitigating the negative impact of graph structural perturbations. To optimize the contrastive learning task, the extracted meta-knowledge was propagate throughout the original graph to learn reliable embedding representations. The self-information learning between views enhances the semantic information of nodes, thus alleviating the problem of over-smoothing. Experimental results on three real-world datasets demonstrate the significant improvement of CollaGCL over state-of-the-art methods.

Urban heat load assessment in Zagreb, Croatia: a multi-scale analysis using mobile measurement and satellite imagery.

A limited number of meteorological stations and sparse data challenge microclimate assessment in urban areas. Therefore, it is necessary to complement these data with additional measurements to achieve a denser spatial coverage, enabling a detailed representation of the city's microclimatic features. In this study, conducted in Zagreb, Croatia, mobile air temperature measurements were utilized and compared with satellite-derived land surface temperature (LST). Here, air temperature measurements were carried out using bicycles and an instrument with a GPS receiver and temperature probe during a heat wave in June 2021, capturing the spatial pattern of air temperature to highlight the city's microclimate characteristics (i.e. urban heat load; UHL) in extremely hot weather conditions. Simultaneously, remotely sensed LST was retrieved from the Landsat-8 satellite. Air temperature measurements were compared to city-specific street type classification, while neighbourhood heat load characteristics were analysed based on local climate zones (LCZ) and LST. Results indicated significant thermal differences between surface types and urban forms and between street types and LCZs. Air temperatures reached up to 35 °C, while LST exceeded 40 °C. City parks, tree-lined streets and areas near blue infrastructure were 1.5-3 °C cooler than densely built areas. Temperature contrasts between LCZs in terms of median LST were more emphasised and reached 9 °C between some classes. These findings highlight the importance of preserving green areas to reduce UHL and enhance urban resilience. Here, exemplified by the city of Zagreb, it has been demonstrated that the use of multiple datasets allows a comprehensive understanding of temperature patterns and their implications for urban climate research.

Predicting the Remaining Useful Life of Lithium-Ion Batteries Using 10 Random Data Points and a Flexible Parallel Neural Network

Accurate Remaining Useful Life (RUL) prediction of lithium batteries is crucial for enhancing their performance and extending their lifespan. Existing studies focus on continuous or relatively sparse datasets; however, continuous and complete datasets are rarely available in practical applications due to missing or inaccessible data. This study attempts to achieve the prediction of lithium battery RUL using random sparse data from only 10 data points, aligning more closely with practical industrial scenarios. Furthermore, we introduce the application of a Flexible Parallel Neural Network (FPNN) for the first time in predicting the RUL of lithium batteries. By combining these two approaches, our tests on the MIT dataset show that by randomly downsampling 10 points per cycle from 10 cycles, we can reconstruct new meaningful features and achieve a Mean Absolute Percentage Error (MAPE) of 2.36% in predicting the RUL. When the input data are limited to the first 10 cycles using the dataset constructed from random downsampling and the FPNN, the predicted RUL MAPE is 0.75%. The method proposed in this study offers an accurate, adaptable, and comprehensible new solution for predicting the RUL of lithium batteries, paving a new research path in the field of battery health monitoring.

An invariance constrained deep learning network for partial differential equation discovery

The discovery of partial differential equations (PDEs) from datasets has attracted increased attention. However, the discovery of governing equations from sparse data with high noise is still very challenging due to the difficulty of derivatives computation and the disturbance of noise. Moreover, the selection principles for the candidate library to meet physical laws need to be further studied. The invariance is one of the fundamental laws for governing equations. In this study, we propose an invariance constrained deep learning network (ICNet) for the discovery of PDEs. Considering that temporal and spatial translation invariance (Galilean invariance) is a fundamental property of physical laws, we filter the candidates that cannot meet the requirement of the Galilean transformations. Subsequently, we embedded the fixed and possible terms into the loss function of neural network, significantly countering the effect of sparse data with high noise. Then, by filtering out redundant terms without fixing learnable parameters during the training process, the governing equations discovered by the ICNet method can effectively approximate the real governing equations. We select the one-dimensional (1D) Kuramoto–Sivashinsky equation, the two-dimensional (2D) Burgers equation, the equation of two-dimensional channel flow over an obstacle, and the equation of three-dimensional (3D) intracranial aneurysm as examples to verify the superiority of the ICNet for fluid mechanics. Furthermore, we extend similar invariance methods to the discovery of wave equation (Lorentz Invariance) and verify it through single and coupled Klein–Gordon equation. The results show that the ICNet method with physical constraints exhibits excellent performance in governing equations discovery from sparse and noisy data.

Dimensions of data sparseness and their effect on supply chain visibility

Supply chain visibility concerns the ability to track parts, components, or products in transit from supplier to customer. The data that organizations can obtain to establish or improve supply chain visibility is often sparse. This paper presents a classification of the dimensions of data sparseness and quantitatively explores the impact of these dimensions on supply chain visibility. Based on a review of supply chain visibility and data quality literature, this study proposes to characterize data sparseness as a lack of data quality across the entire supply chain, where data sparseness can be classified into three dimensions: noise, bias, and missing values. The quantitative analysis relies on a stylized simulation model of a moderately complex illicit supply chain. Scenarios are used to evaluate the combined effect of the individual dimensions from actors with different perspectives in the supply chain, either supply or demand-oriented. Results show that when a data sparseness of 90% is applied, supply chain visibility reduces to 52% for noise, to 65% for bias, and to 32% for missing values. The scenarios also show that companies with a supply-oriented view typically have a higher supply chain visibility than those with a demand-oriented view. The classification and assessment offer valuable insights for improving data quality and for enhancing supply chain visibility.

Supporting ANFIS interpolation for image super resolution with fuzzy rough feature selection

Image Super-Resolution (ISR) is utilised to generate a high-resolution image from a low-resolution one. However, most current techniques for ISR confront three main constraints: i) the assumption that there is sufficient data available for training, ii) the presumption that areas of the images concerned do not involve missing data, and iii) the development of a computationally efficient model that does not compromise performance. In addressing these issues, this study proposes a novel lightweight approach termed Fuzzy Rough Feature Selection-based ANFIS Interpolation (FRFS-ANFISI) for ISR. Popular feature extraction algorithms are employed to extract the potentially significant features from images, and population-based search mechanisms are utilised to implement effective FRFS methods that assist in selecting the most important features among them. Subsequently, the processed data is entered into the ANFIS interpolation model to execute the ISR operation. To tackle the sparse data challenge, two adjacent ANFIS models are trained with sufficient data where appropriate, intending to position the ANFIS model of sparse data in the middle. This enables the two neighbouring ANFIS models to be interpolated to produce the otherwise missing knowledge or rules for the model in between, thereby estimating the corresponding outcomes. Conducted on standard ISR benchmark datasets while considering both sufficient and sparse data scenarios, the experimental studies demonstrate the efficacy of the proposed approach in helping deal with the aforementioned challenges facing ISR.

A Novel Ionospheric Inversion Model: PINN‐SAMI3 (Physics Informed Neural Network Based on SAMI3)

AbstractPurely data‐driven ionospheric modeling fails to adequately obey fundamental physical laws. To overcome this shortcoming, we propose a novel ionospheric inversion model, Physics‐Informed Neural Network based on fully physical models SAMI3 (PINN‐SAMI3). The model incorporates the governing equations of the ionospheric physical model SAMI3 into the neural network to reconstruct the temporal‐spatial distribution of ionospheric plasma parameters. The objective of this study is to investigate the feasibility of integrating physical models with machine learning for ionospheric modeling. The PINN‐SAMI3 framework enforces physical laws through the multiple ion species of continuity, momentum, temperature equations in the magnetic dipole coordinate system. The simulation results show that if sparse ion densities are used as training data, it is possible to retrieve ionospheric electron densities, ion velocities and ion temperatures, respectively. The optimal physical constraints have been also investigated for different inversion quantities. Furthermore, the impact of incorporating E × B velocity terms on inversion results during the periods of ionospheric calm and geomagnetic storm is analyzed. The PINN‐SAMI3 achieves good inversion results even using sparse data in comparison to the traditional artificial neural networks (ANN). The framework will contribute to advance the future space weather prediction capability with artificial intelligence (AI).

Cosmic Sands. II. Challenges in Predicting and Measuring High-z Dust Temperatures

In the current era of high-z galaxy discovery with JWST and the Atacama Large Millimeter/submillimeter Array, our ability to study the stellar populations and interstellar medium conditions in a diverse range of galaxies at Cosmic Dawn has rapidly improved. At the same time, the need to understand the current limitations in modeling galaxy formation processes and physical properties in order to interpret these observations is critical. Here, we study the challenges in modeling galaxy dust temperatures, both in the context of forward modeling galaxy spectral properties from a hydrodynamical simulation and via backwards modeling galaxy physical properties from mock observations of far-infrared dust emission. Using the simba model for galaxy formation combined with powderday radiative transfer, we can accurately predict the evolution of dust at high redshift, though several aspects of the model are essentially free parameters (dust composition, subresolution dust in star-forming regions) that dull the predictive power of the model dust temperature distributions. We also highlight the uncertainties in the backwards modeling methods, where we find the commonly used models and assumptions to fit far-infrared spectral energy distributions and infer dust temperatures (e.g., single temperature, optically thin modified blackbody) largely fail to capture the complexity of high-z dusty galaxies. We caution that conclusions inferred from both simulations—limited by resolution and post-processing techniques—and observations—limited by sparse data and simplistic model parameterizations—are susceptible to unique and nuanced uncertainties that can limit the usefulness of current high-z dust measurements.

Reconstructive Classification for Age and Gender Identification in Social Networks

Author profiling is a challenging task that consists of identifying a person’s relevant attributes based on the content he/she generates. In this article, we validate a classification method based on the reconstructive classification to identify two demographic attributes, age and gender, for users of social networks based on the text content they publish. For the problem, we consider balanced and unbalanced data, where a gender or age group has a larger presence than the others. The proposed method is based on using the reconstructive property of singular value decomposition (SVD), and its suitability to work with sparse data, to find a matrix with the main latent components to represent the information of specific gender or age classes. Afterward, we use such a matrix to project and reconstruct new users’ information to identify their demographic variables. We test our method with a set of datasets in several languages collected from Twitter and Pinterest users, and we use different evaluation metrics to compare their performance with the ones of several popular classifiers based on traditional machine learning, and on deep learning, and with some relevant works in the literature. The results show that the proposed classifier performs generally well in identifying the age and gender of users in social networks.