Sistema Regional de Información
en línea para Revistas Científicas de América Latina,
el Caribe, España y Portugal

This article reflects on graphic representations of the face in the digital environment through an analysis of digital avatars known as memojis, which have evolved from emojis by allowing the user's face to be personalized. Unlike conventional emojis, which express emotions in a standardized way, memojis make it possible to create customized avatars, incorporating the individual's physical characteristics. In addition, they can be used not only in text messages, but also to replace profile photos or faces in video interactions. Using a theoretical approach to digital culture and contemporary subjectivity, this study examines Apple's memojis through the lens of Peircean semiotics. The hypothesis guiding the work suggests that replacing the real face with a memoji could enhance expressiveness and individuality, although it also carries risks of homogenization. We conclude that these figures can be used as masks in increasingly sophisticated communication practices, but still in a limited manner, since they are based on an essentially optimistic perspective in a somewhat idealized world.  

This article seeks to establish a reflective dimension regarding the conditions and problems involved in facial modeling. To do so, it will utilize theoretical considerations from cognitive semiotics and, at the same time, illustrate the reflection by considering some relevant cases. Specifically, this article seeks to raise specific questions regarding Kant's (2010) approach to the interpretive function of the face: for the German philosopher, the face and the gaze allow us to intuit a subject's temperament. Thus, the theoretical and methodological foundations of cognitive semiotics will be used to argue Kant's proposal and, at the same time, establish the cognitive conditions involved in facial modeling.

Interview with Letícia Cesarino.

Interview with Sandra Savoini.

Studies on the face have a long tradition in fields such as anthropology, philosophy, art, semiotics, and computer science. What meanings do facial features confer on bodies? When does the transition from a face to a visage occur? What are the minimal phenomenological requirements for the perception of a face to emerge? When does a face become “friendly,” “recognizable,” or “identifiable”?

This study presents a machine learning framework for multiclass classification of neurological and psychiatric disorders using synthetic neuroinformatics biomarkers and EEG spectral simulations. Synthetic data modeled cognitive-motor features (visual delay, motor gain, expectancy weight, memory capacity, EEG bands), while real EEG data from 50 PhysioNet recordings validated performance. An XGBoost model optimized through grid search (24 combinations, five-fold validation) achieved 97.8% accuracy and 0.978 F1-score. Results showed excellent classification of neurotypical, ADHD, ASD, dementia, depression, GAD, Parkinson’s, psychosis, and Tourette’s, demonstrating strong potential for neuropsychiatric diagnosis support.

The analysis of genomic data allows to comprehend biological processes at the molecular level. A challenging application is the classification of antibodies according to the antigens they bind. Antibodies, the heart of the immune system, are proteins that bind to specific antigens to inactivate pathogens. Antibody classification requires datasets with structural and functional information about antibodies. The Observed Antibody Space is a dataset collecting genomic sequences of antibodies from several species and OAS contains information on the structure of antibodies. This paper examines the impact of dimensionality reduction techniques on the classification of SARS-CoV-2 antibodies, utilizing genetic sequence data from the Observed Antibody Space database. Specifically, we focus on transforming amino acid sequences from the Complementarity Determining Region, CDR, into word embeddings for subsequent processing in machine learning models. This transformation enables the use of unlabeled, high-dimensional data but presents the challenge of the curse of dimensionality, which can affect the models’ accuracy and efficiency. To address this problem, two dimensionality reduction techniques are applied and evaluated: Principal Component Analysis and Uniform Manifold Approximation and Projection. We developed 36 classification models using Support Vector Machines, Random Forests, and K-Nearest Neighbors algorithms, testing each on original datasets and on reduced datasets. The objective is to determine whether dimensionality reduction improves model performance. The study provides insights into how these techniques can facilitate predictive analysis in bioinformatics and contribute to the development of efficient models for identifying relevant antibodies in immunology.

The Capacitated Vehicle Routing Problem (CVRP) consists of generating a customer route for each vehicle in which the sum of the customer demands does not exceed the vehicle's capacity. Each vehicle must start from the depot, and when it finishes visiting the last customer, it must return to the depot, thus managing to visit all customers only once by any of the vehicles and have a route solution where the sum of the distances of all the routes is the minimum. This research proposed using randomness and the Swap local search algorithm to create an initial solution with a good neighborhood. Later, the Tabu Search algorithm is employed to explore the solution space and obtain an improved solution. The proposed algorithms give feasible solutions with an approximation of less than a 10% difference, and in some cases, the best-known solution is obtained.  

Power transformers are equipment of great importance, and their availability is crucial for the security and continuity of the electricity supply for domestic and industrial users. During their life cycle, transformers are exposed to various environmental and operational conditions that affect their performance, especially when these exceed the operational design limits. This paper describes the use of Fuzzy Logic models as supporting tools for the automatic classification of power transformer operating conditions. The proposed methodology involves a binary classification (failure or no failure), followed by a multi-classification into seven types of failures. For this purpose, a power transformer fault database was developed, compiling information from operational data curated by power transformer experts. The results show a high predictive capacity for transformer fault conditions, with 96% balanced accuracy, and acceptable effectiveness in detecting different faults. This approach may serve as useful guidance in power transformer condition monitoring, helping engineers to reduce the time required to detect and repair incipient faults.

This paper examines the continuum from natural to artificial consciousness, highlighting the biological basis of subjective experience. It introduces synthetic consciousness as a hybrid of neural and algorithmic systems and examines ethical, legal, and ontological implications of human–AI integration. The Pyramid of Consciousness framework guides reflection on autonomy, identity, and the shifting boundary between organic cognition and intelligent machines.