Conference Agenda

Controlling the costs associated with corporate healthcare utilization and employee absenteeism represents an increasing challenge for organizational sustainability, particularly in emerging economies. This study applies a People Analytics framework to examine the relationship between medical service utilization, work incapacity, and associated economic losses using monthly aggregated administrative data. The methodology combines exploratory data analysis, predictive modeling through Ridge regression with temporal validation, scenario analysis, and pre–post comparisons. Results reveal substantial temporal variability and a strong concentration of costs among a limited number of healthcare providers. The predictive model significantly outperforms a naïve temporal baseline, highlighting the importance of service utilization volume and temporal persistence of incapacity. In addition, the comparison between July and November 2025 shows reductions exceeding 30% in both incapacity hours and total economic loss. Overall, the findings demonstrate the value of People Analytics as an evidence-based tool to support preventive decision-making and enhance financial and operational sustainability.

In this article, the combination of software tools such as TIA Portal, PLC, LabVIEW, OPC Server, and MATLAB was used to address the challenge of vehicular traffic control in a simulated environment. Therefore, the objective was to develop a fuzzy inference algorithm implemented in a programmable logic controller, with the support of a Matlab software Simulink library, to control vehicular traffic at the intersection of two main avenues in the Santiago de Surco district, in Lima, Peru. To do this, the LabVIEW programming environment was used, which allowed the creation of a graphical interface capable of simulating vehicular traffic, as close to real conditions as possible. This simulator was based on observable data collected from the field that allowed the generation of various traffic scenarios. In addition, the MATLAB Fuzzy Logic toolbox was used to implement the fuzzy inference algorithm, based on the observed conditions. Then, a PLC and TIA Portal software were used to supervise and simulate control of the traffic lights and traffic sensors. Likewise, communication between LabVIEW, MATLAB and the PLC was achieved using the OPC Server, which allowed the transfer of data and commands between these platforms efficiently. In this way, this proposal was tested in a virtual laboratory environment, obtaining significant results in terms of traffic fluidity, and a reduction in congestion times at the analyzed intersection.

This paper details the development and implementation of an artificial intelligence system for the automatic detection of liver cirrhosis. Given the subjectivity of traditional ultrasound diagnosis, a computer vision strategy based on Deep Learning is proposed. The methodology included the preprocessing of ultrasound images, applying spatial rescaling techniques to 150x150 pixels and intensity normalization [0,1] prior to network input. A custom 7-layer Convolutional Neural Network (CNN) was designed and trained on Google Colab, achieving a validation accuracy of 92.00% and a perfect sensitivity of 1.00 for the pathological class. To ensure clinical applicability, the model was deployed as a cloud microservice using Docker and FastAPI on Hugging Face, consumed by a native Android mobile application (Kotlin). Results validate the technical feasibility of using lightweight architectures to democratize access to diagnostic tools.

Youth depression and suicide represent an increasingly critical public health challenge in Peru, with a significant impact on the university population. This paper presents a technological model based on an emotional-support chatbot that integrates the GPT-4 architecture and Natural Language Processing (NLP) techniques. The approach combines sentiment analysis with the generation of empathetic responses, complemented by the delivery of educational resources. Preliminary results indicate that the chatbot provides a safe and accessible space for emotional support. This proposal aims to strengthen preventive mental health strategies in the Peruvian university context through the use of innovative digital tools.

Emotional violence in digital environments poses a growing risk in Latin American contexts, where verbal expressions vary significantly by region. This study presents an emotion-detection model based on a Bidirectional Long Short-Term Memory (Bi-LSTM) network, trained specifically on messages collected from Twitter using web-scraping techniques. Unlike prior approaches, the model does not rely on surveys or corpora structured in neutral Spanish; instead, it draws on real, everyday language with contextual emotional content. The methodological pipeline spans from data collection and cleaning to the binary classification of violent versus nonviolent messages. The model achieved an accuracy close to nine out of ten correct predictions in identifying violent messages, with recall similarly high for both classes and an F1 score near 0.9. In addition, the model exhibited semantic capabilities to interpret ambiguous phrases, local expressions, and neutral messages, reducing false positives.

The enrollment process in public schools is an important procedure for school administration. In Northern Lima, most of the enrollment forms are filled manually, which leads to inefficiencies, risk of losing information and administrative overload [1], [2]. This research proposes a technological model based on machine learning techniques aimed at improving the management of these forms. The machine learning model will adapt predictive and classification algorithms to predict how likely is that a new enrollment process will encounter problems or delays due to paperwork required. Recent studies have shown that the right selection of algorithms based on the characteristics of data is critical to achieve effective models on educational and business scenarios [8]. Additionally, international studies have highlighted the importance of explainable machine learning models in predicting the academic performance of high school students, which reinforces the pertinence of applying this scope on public education [7]. Unlike traditional methods, this approach leverages the features of handling large volumes of data and learn from historical patterns, thereby assisting informed decision-making. The study contributes to the digital transformation of the public education system in Peru, with a high potential for scalability at regional and national levels.