Producción científica y docente
Producción científica reciente
Eco-Anxiety in Latin American Citizens: Description and Factors Associated With Concern About Climate Change
Background: Climate change is an undeniable reality that is increasingly discussed globally. While eco-anxiety has been extensively studied in Europe and other developed countries, there is a lack of data from Latin America. Aims: To identify and describe the factors associated with eco-anxiety in Latin America. Method: A cross-sectional, multicenter study used a validated scale to measure eco-anxiety, adjusted for sex, age, education level, and country of residence. Descriptive and analytical statistics were employed to interpret the data. Among 3,320 participants, a significant portion expressed eco-anxiety, particularly when it was mentioned to affect an important place, interfere with concentration, disrupt work or studies, or cause difficulty sleeping (12%, 5%, 3%, and 3% strongly agreed with these statements, respectively). Results: Multivariate analysis revealed higher scores of eco-anxiety in Argentina (aRP: 2.06; 95% CI [1.45, 2.94]; p < 0.001), Colombia (aRP: 1.66; 95% CI [1.21, 2.29]; p = 0.002), and Ecuador (aRP: 1.40; 95% CI [1.01, 1.94]; p = 0.042). Conversely, eco-anxiety was lower among men (aRP: 0.88; 95% CI [0.81, 0.96]; p = 0.003) and those with postgraduate education (aRP: 0.72; 95% CI [0.59, 0.88]; p = 0.001), after adjusting for age. Conclusions: This is the first large-scale study to report significant eco-anxiety in Latin America, which is linked to factors such as country of residence, sex, and educational level.
Framework to Detect Adulterated Honey Using Hyperspectral Imaging and Machine Learning
Honey is a natural product renowned for its nutritional and medicinal properties. However, the increasing market demand in Pakistan has led to frequent instances of adulterated and fraudulent honey. Conventional detection methods are widely used but are often ineffective in accurately identifying adulteration. This study proposes a novel approach to detecting adulteration in honey, specifically the mixing of sugar, using machine learning (ML) techniques in conjunction with hyperspectral imaging (HSI). By leveraging HSI, we capture distinct spectral features of both pure and adulterated honey. These spectral features are processed and analyzed using ML algorithms trained on a dataset comprising pure and mixed honey samples. Experiments involve binary, as well as, multiclass (5 classes) honey samples with different levels of adulteration. In addition, data balancing is also carried out using synthetic minority oversampling technique (SMOTE). The results prove the proposed approach to be efficient in detecting adulteration across various honey varieties, with linear regression achieved 99.9% while support vector machine and multilayer perceptron obtaining a 98.7% accuracy for binary class. For the multiple classes of honey, the multilayer perceptron shows a 96.67% accuracy outperforming the remaining ML and deep learning models. Among the deep learning models, the convolutional neural network (CNN) achieved the best performance with an accuracy of 94.67% after applying the SMOTE × 3 (three times of over samples of original samples of each class) data balancing strategy, whereas recurrent neural network (RNN), long short-term memory (LSTM), and gated recurrent unit (GRU) models showed comparatively lower performance.
Development of the Semi-Implicit Self-Report to assess Attitudes Toward Immigration (SISRATI): Psychometric properties and convergence
Attitudes toward immigration in Spain represent a pressing social concern, particularly in the context of increasing reports of racism and public rejection and concern. However, assessing these attitudes is complicated by social desirability bias inherent in self-report measures. This study aimed to develop and validate a semi-implicit instrument using news headlines—half pro-immigration and half anti-immigration—which participants (N = 616, divided into two sub-samples) judged as true or false. Psychometric properties were evaluated through Exploratory and Confirmatory Factor Analyses, revealing a two-factor structure comprising 12 items: “Pro-Immigration Statements” and “Anti-Immigration Statements.” Based on these dimensions, participants were clustered into four groups: “Favorable to Immigration,” “Unfavorable to Immigration,” “Non-Discriminant Believers,” and “Non-Discriminant Unbelievers.” These groups were then compared on emotional responses (valence and intensity) to immigration-related content, Big Five personality traits, and reflective cognitive style. Significant differences in emotional responses across groups supported the tool’s convergent validity. The results highlight the utility of semi-implicit measures in capturing nuanced attitudes toward immigration and avoiding the biases of traditional self-report methods. Implications for future research and social intervention are discussed, along with study limitations and potential improvements for the proposed tool.
Cryptocurrency Market Trends: A Machine Learning-Driven Time Series Forecasting with Twitter Sentiment Integration
Accurate forecasting of cryptocurrency prices remains an open challenge because classical statistical models cannot capture the non-linear, sentiment-driven dynamics of these markets. This study compares three hybrid deep learning architectures—VAR-LSTM, XGBoost-LSTM, and CNN-LSTM—to determine which best forecasts Bitcoin (BTC), Ethereum (ETH), and Dogecoin (DOGE) closing prices, and to quantify the marginal predictive value of Twitter sentiment integration. Six years of hourly OHLCV data (2017–2023) are augmented with VADER-scored Twitter sentiment polarity. Each model is formulated mathematically, implemented with documented hyperparameters (epochs, dropout, units;), and trained for one-step-ahead next-hour price prediction. Performance is measured by RMSE, MAE, MAPE, R 2 , and Directional Accuracy (DA) across five random seeds, with paired Wilcoxon significance tests. XGBoost-LSTM achieves the best performance (RMSE = 81.547, R 2 = 0.9254, DA = 80.0%), outperforming all nine literature baselines. Removing Twitter sentiment degrades DA by 14.3 percentage points ( p < 0.01 ), confirming that social media signals carry independent predictive information. Hybrid architectures consistently outperform single-model baselines; XGBoost-LSTM offers the best accuracy-to-compute ratio. VADER-enriched Twitter sentiment is a significant predictor beyond price history. Limitations include reliance on a single sentiment platform and a training window that predates several structural market events
Validation of the CONSULT-PSYCHIATR Scale: reasons for reluctance to seek psychiatric consultation among workers in Peru
Introduction: Workers may be reluctant to attend psychiatric consultation because of stigma, interpersonal concerns, and practical barriers. In Peru, brief instruments specifically designed to assess these reasons in working populations are limited. This study examined the content validity, internal structure, and internal consistency of a Peruvian scale assessing reasons why workers are reluctant to attend consultations with a psychiatrist.Methods: An instrumental study was conducted in Peru. Item development was informed by a literature review and expert input. An initial 10-item version was evaluated by 20 experts for relevance, representativeness, and clarity using Aiken’s V. After a pilot test, the questionnaire was administered to 3001 workers from several Peruvian departments using non-random sampling. Item distribution was examined using descriptive statistics, skewness, excess kurtosis, floor and ceiling effects, and constant-response patterns. The sample was randomly divided into exploratory and confirmatory subsamples. A polychoric exploratory factor analysis was conducted in the exploratory subsample, followed by confirmatory analyses in the independent confirmatory subsample. Internal consistency was estimated using Cronbach’s alpha, ordinal alpha, and ordinal omega total.Results: Most Aiken’s V coefficients were 0.70 or higher, with only one lower value for the representativeness of item 9. The 10-item version showed adequate item distributions, although 46.95% of participants provided constant responses across all items, prompting sensitivity analyses. Exploratory factor analysis supported a one-factor structure, explaining 80.96% of the variance, with factor loadings ranging from 0.8425 to 0.9365. This structure remained stable after excluding constant-response patterns. In the confirmatory subsample, the 10-item one-factor model showed high standardized loadings and favorable CFI, TLI, and SRMR values, although RMSEA was less favorable and was interpreted jointly with the remaining indices. Reliability estimates were high in both the main and sensitivity analyses.Discussion: The CONSULT-PSYCHIATR is a brief 10-item instrument with evidence of content validity, internal structure, and internal consistency for assessing perceived reasons why workers in Peru may be reluctant to attend psychiatric consultation. Although the findings are encouraging, further studies are needed to examine its performance in other populations and settings and to assess additional sources of validity evidence.