Depression is regarded as a widespread mental condition that affects people of all ages. It has a negative impact on a variety of aspects of life, including mood, vigor, and interests in enjoying activities. In the most severe cases, depression can also result in suicide. creating the chance for collaboration between mental health professionals and the use of technical tools to enhance the assessment of the severity of depression to offer the patient with an ideal clinical diagnosis and an appropriate referral to begin treatment. The COVID-19 epidemic in Peru has decreased face-to-face interaction and quick access to medical professionals, making it more difficult for patients’ mental health to be identified or treated effectively, which results in the disease becoming chronic, psychological suffering, and high costs associated with specialized care. The implementation of a technology model that assesses degrees of recurrent depression by examining facial photos and voice to identify the chronicity of depressive symptoms in young Peruvians is thus one of the research’s problems. Our findings demonstrate that, based on the functions of the mobile application, adolescent patients were predisposed to complete a self-administered depression questionnaire in a simulated setting with an optimal feeling of satisfaction and usefulness.

In this paper, we present a novel system that combines process discovery and conformance checking to optimize resource usage in the operations area of banking companies. Our proposed system improves upon previous solutions by integrating real-time process mining techniques to reduce resource consumption in banking operations. The use of real-time process mining techniques in the proposed system enables faster error identification in processes, emphasizing the importance of such techniques for banking companies. The system is comprised of two integrated process mining techniques, as reported in existing literature. Process discovery involves two key components: dataset generation and process visualization. Additionally, compliance checking involves process monitoring, alerts, and reports. To validate the effectiveness of our system, we conducted a stress test, which was evaluated by experienced process mining users with extensive experience in the banking sector in Lima, Peru. Preliminary results of the stress test demonstrated a significant improvement in system performance, with a 37.42% increase in the capacity of web requests per minute and a total of 82.86% of requests successfully answered. These results enabled corrective and preventive actions to be taken, indicating the practical value of our proposed system in optimizing resource usage in banking operations. In conclusion, our system provides a robust solution to optimize resource usage in the operations area of banking companies and emphasizes the importance of real-time process mining techniques in achieving this goal. The system has been validated through a stress test and evaluation by experienced users in the banking sector, further indicating its practical applicability.

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado.

The information's management on the traceability of wood in Peru is complicated by multiple causes. In this article we present our project, which seeks to propose a technological model that allows greater efficiency in the management of timber information so that sawmills in Peru can improve their administration. Our proposed model allows us to appreciate a possible improvement in the time it takes to obtain information on the traceability of wood compared to the traditional way in which work continues in Peru. Also, in the investigation a prototype application of traceable information was developed. The main contribution of the research consists in the design of a technological model that serves as a reference to develop a technological solution for the traceability of wood in Peru.

Visually impaired people experience different accessibility barriers when they want to learn a second language through virtual classes. In addition to this, Screen Readers often have a high learning curve, plus they are not optimized for reading different languages at the same time. For this reason, it was decided to design the model of a Learning Management System (LMS) with a minimalist and accessible design and with an integrated voice assistant. As a result, it was obtained that the proposed solution was more efficient and effective in providing support to the process of teaching foreign languages in contrast to the methodology currently used.

so, machine learning techniques are being developed to improve performance and maintenance prediction. Increasing our knowledge of the relationship between humans and algorithms, Because data is so valuable, improving strategies for intelligently having to manage the now-ubiquitous content infrastructures is a necessary part of the process toward completely autonomous agents. Numerous researchers recently developed numerous computer-aided diagnostic algorithms employing various supervised learning approaches. Early identification of sickness may help to reduce the number of people who die as a result of these illnesses. Using machine learning techniques, this research creates an efficient automated illness diagnostic algorithm. We chose three key disorders in this paper: coronavirus, cardiovascular diseases, and diabetes. The data are inputted into a mobile application in the suggested model, the investigation is then done in a real-time dataset that used a pre-trained model machine learning technique trained within the same dataset then implemented in firebase, and lastly, the illness identification result can be seen in the mobile application. Logistic regression is a method of prediction calculation

SMEs in the textile sector face many problems in their production flows, mainly due to the lack of production management systems caused by poor management of the production chain. Therefore, a diagnostic analysis is carried out in a textile SME to evaluate and define the deficiencies and factors that affect its competitiveness, which began with the analysis of the current situation of the company, where it was established the existence of poor-quality management, high waiting times and lack of procedures. Therefore, the use of lean manufacturing tools such as Jidoka, Single Minute Exchange of Die with respect to the production line and process management for the measurement and control of operations in the production area is proposed. The incorporation of these tools in block allows to decrease the rates of defective products, the excess of operative work and the set-up of the machines for the change of model. The main result of the research was that production increased to 0.091 und/PEN.

Many homes in the country of Peru, especially those located in unofficial settlements, are not connected to public service networks, and in the case of residential water, require tanker truck delivery. However, this water has often been contaminated from the upstream storage, conveyance and delivery systems that provide it, and thus will not comply with government water quality standards, ultimately compromising the health of the people who rely on it. While the topic of quality monitoring in traditional water networks has been studied, research has not focused on water quality control in under-developed and under-served unofficial settlements. This study introduces an IoT architecture and web-based system for real-Time monitoring of the key water quality parameters to help municipalities and other government entities to act early when large volumes of low-quality water are detected. The system proposed was implemented across five layers: capture, communication, processing, storage and presentation. Two experiments were conducted in a residential home with real time measurement of temperature, turbidity, TDS y pH. When comparing the results of both experiments, the pH parameter had a better precision with a 2% error rate. In addition, the survey results showed that the experts agree with the proposal.

Deep learning (DL) is a new approach that provides exceptional speed in healthcare activities with greater accuracy. In this regard, "convolutional neural network"or CNN and blockchain are two important parts that together fasten the disease detection procedures securely. CNN can detect and predict diseases like lung cancer and help determine food quality, and blockchain is responsible for data. This research is going to analyze the extension of blockchain with the help of CNN for lung cancer prediction and making food safer. CNN algorithm has been trained with a huge number of images by altering the filters, features, epoch values, padding value, kernel size, and resolution. Subsequently, the CNN accuracy has been measured to understand how these factors affect the accuracy. A linear regression analysis has been carried out in IBM SPSS where the independent variables selected are image dataset augmentation, epochs, features, pixel size (90 × 90 to 512 × 512), kernel size (0-7), filters (10-40), and padding. The dependent variable is the accuracy of CNN. Findings suggested that a larger number of epochs improve the CNN accuracy; however, when more than 12 epochs are considered, the accuracy may decrease. A greater pixel/resolution also improves the accuracy of cancer and food image detection. When images are provided with excellent features and filters, the CNN accuracy improves. The main objective of this research is to comprehend how the independent variables affect the accuracy (dependent), but the reading may not be fully exact, and thus, the researcher has conceded out a minor task, which delivered evidence supportive of the analysis and against the analysis. As a result, it can be determined that image augmentation and a large number of images develop the CNN accuracy in lung cancer prediction and food safety determination when features and filters are applied correctly. A total of 10-12 epochs are desirable for CNN to receive 99% accuracy with 1 padding.

With the increasing popularity of online social networking platforms, the amount of social data has grown exponentially. Social data analysis is essential as spamming activities and spammers are escalating over online social networking platforms. This paper focuses on spammer detection on the Twitter social networking platform. Although existing researchers have developed numerous machine learning methods to detect spammers, these methods are inefficient for appropriately detecting spammers on Twitter due to the imbalance of spam and nonspam data distribution, the involvement of diverse features and the applicability of data mechanisms by spammers to avoid their detection. This research work proposes a novel hybrid approach of the gravitational search algorithm and the decision tree (HGSDT) for detecting Twitter spammers. The individual decision tree (DT) algorithm is not able to address the challenges as it is unstable and ineffective for the higher level of favorable data for a particular attribute. The gravitational search algorithm (GSA) constructs the DTs with improved performance as the gravitational forces act as the information-transferring agents through mass agents. Moreover, the GSA is efficient in handling the data of higher dimensional search space. In the HGSDT approach, the construction of the DT and splitting of nodes are performed with the heuristic function and Newton's laws. The performance of the proposed HGSDT approach is determined for the Social Honeypot dataset and 1KS-10KN dataset by conducting three different experiments to analyze the impact of training data size, features and spammer ratio. The result of the first experiment shows the need of a higher proportion of training data size, the second experiment signifies the more importance of textual content-based features compared to the other feature categories and the third experiment indicates the requirement of balanced data to attain the effective performance of the proposed approach. The overall performance comparison indicates that the proposed HGSDT approach is superior to the incorporated machine learning methods of DT, support vector machine and back propagation neural network for detecting Twitter spammers.