RESUMO
In clinical and scientific research on emotion recognition using physiological signals, selecting the appropriate segment is of utmost importance for enhanced results. In our study, we optimized the electrodermal activity (EDA) segment for an emotion recognition system. Initially, we obtained EDA signals from two publicly available datasets: the Continuously annotated signals of emotion (CASE) and Wearable stress and affect detection (WESAD) for 4-class dimensional and three-class categorical emotional classification, respectively. These signals were pre-processed, and decomposed into phasic signals using the 'convex optimization to EDA' method. Further, the phasic signals were segmented into two equal parts, each subsequently segmented into five nonoverlapping windows. Spectrograms were then generated using short-time Fourier transform and Mel-frequency cepstrum for each window, from which we extracted 85 features. We built four machine learning models for the first part, second part, and whole phasic signals to investigate their performance in emotion recognition. In the CASE dataset, we achieved the highest multi-class accuracy of 62.54% using the whole phasic and 61.75% with the second part phasic signals. Conversely, the WESAD dataset demonstrated superior performance in three-class emotions classification, attaining an accuracy of 96.44% for both whole phasic and second part phasic segments. As a result, the second part of EDA is strongly recommended for optimal outcomes.
Assuntos
Emoções , Resposta Galvânica da Pele , Emoções/fisiologia , Aprendizado de MáquinaRESUMO
The human gut microbiome contributes to a broad range of biochemical and metabolic functions that directly or indirectly affect human physiology. Several recent studies have indicated that factors like age, geographical location, genetic makeup, and individual health status significantly influence the diversity, stability, and resilience of the gut microbiome. Of the mentioned factors, geographical location (and related dietary/socio-economic context) appears to explain a significant portion of microbiome variation observed in various previously conducted base-line studies on human gut microbiome. Given this context, we have undertaken a microbiome study with the objective of cataloguing the taxonomic diversity of gut microbiomes sampled from an urban cohort from Ahmedabad city in Western India. Computational analysis of microbiome sequence data corresponding to 160 stool samples (collected from 80 healthy individuals at two time-points, 60 days apart) has indicated a Prevotella-dominated microbial community. Given that the typical diet of participants included carbohydrate and fibre-rich components (predominantly whole grains and legume-based preparations), results appear to validate the proposed correlation between diet/geography and microbiome composition. Comparative analysis of obtained gut microbiome profiles with previously published microbiome profiles from US, China, Finland, and Japan additionally reveals a distinct taxonomic and (inferred) functional niche for the sampled microbiomes.