Ciencias Exactas y Ciencias de la Salud
Permanent URI for this collectionhttps://hdl.handle.net/11285/551039
Pertenecen a esta colección Tesis y Trabajos de grado de las Maestrías correspondientes a las Escuelas de Ingeniería y Ciencias así como a Medicina y Ciencias de la Salud.
Browse
Search Results
- Unsupervised learning to profile emerging researchers in LATAM with Elsevier’s data(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2024-05) Figueroa Castillo, Jesús Manuel; Hernández Gress, Neil; mtyahinojosa, emimmayorquin; Ceballos Cancino, Héctor Gibrán; Estévez Bretón, Carlos Manuel; School of Engineering and Sciences; Campus Monterrey; Hervert Escobar, LauraThis proposal is being presented in Computer Science. High-impact researchers possess several key features based on their expertise; never theless, it takes time to establish themselves as leaders in their area. The objective of this research is to develop a model that can identify those outstanding researchers by discipline using indicators from the last five years of research and acknowledged databases such as Sco pus and Web of Science. Additionally, it will compare similarities across various disciplines to determine whether it is possible to predict researchers from one or more disciplines using the same model. The main objective of this research is to discover the characteristics that define a ”rising star” based on the concept of an early career researcher as a initial time window. It is important to mention that current metrics measure researchers’ performance through indicators known as H-index and its variants. However, these metrics often do not consider characteristics that differentiate one group from another. Through this unsupervised approach, we aim to f ind different groups that exist in LATAM to measure their characteristics more precisely and fairly, and to identify those high-impact researchers who may not be immediately apparent through indicators like the H-index. This thesis will demonstrate the process from data mining to the statistical analysis of the different groups.
- Evolutionary clustering using classifiers: definition, implementation, scalability, and applications(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-04-22) Sainz Tinajero, Benjamin Mario; GUTIERREZ RODRIGUEZ, ANDRES EDUARDO; 436765; Guiterrez Rodriguez, Andres Eduardo; puemcuervo; Ortiz Bayliss, Jose Carlos; Amaya Contreras, Ivan Mauricio; Medina Perez, Miguel Angel; School of Engineering and Sciences; Campus Estado de MéxicoClustering is a Machine Learning tool for partitioning multi-dimensional data automatically into mutually exclusive groups, aiming to reflect the patterns of the phenomena it represents. Clustering algorithms perform this task conditioned by the clustering criterion modeled in its objective function. However, selecting the optimal criterion is a domain-dependent task that requires information on the cluster structure that a user often does not count on due to the unsupervised nature of the technique. Available approaches accentuate this problem as they perform clustering according to a similarity notion often limited to the concepts of compactness and connectedness, inducing bias and favoring clusters with certain shape, size, or density properties from using conventional distance functions. However, we cannot consider this a complete notion of a cluster because not every dataset will comply with both notions in the same proportion. Hence, research on this topic has not converged to a standard definition of a cluster, which raises the need for algorithms that produce adaptive solutions that mirror the underlying structures and relations within the data. This thesis is focused on the design of single-objective Evolutionary Clustering Algorithms that generate solutions that are not biased towards any cluster structure by optimizing a novel generalization clustering criterion. To achieve that, we designed objective functions modeled as a supervised learning problem, considering that a good partition should induce a well-trained classifier. That is how we decided to assess the quality of a clustering solution, according to its capability to train an ensemble of classifiers. The main contribution of this thesis is our series of Evolutionary Clustering Algorithms using Classifiers (the ECAC series), which introduces the aforementioned clustering criterion along with evolutionary computation. This meta-heuristic allows us to model distinct criteria to optimize while creating and evaluating multiple solutions along the process. The experimental results in the design of our family of methods ECAC, F1-ECAC, and ECAC-S, show an increase in similarity between the partitions created by our algorithms and the ground truth labels (obtained from the publicly available repositories where we retrieved the data) with a maximum Adjusted RAND Index of 0.96. Our second algorithm, F1-ECAC, proved the competitiveness of our contributions against traditional, single, and multi-objective Evolutionary Clustering algorithms showing no statistically significant difference against k-means, HG-means, and MOCLE. Our latest contribution, ECAC-S, was tested on a satellite image segmentation task, and it produced segmentations with higher average Adjusted RAND Index than k-means, Spectral-clustering, Birch, and DBSCAN in 4 out of 10 images.
- One step closer to mental health: resilience to mental stress index in the face of analytical problems, a machine learning approach(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2021-06-01) Díaz Ramos, Ramón Eduardo; Trejo Rodríguez, Luis Angel; puemcuervo; Medina Pérez, Miguel Angel; González Mendoza, Miguel; Figueroa López, Carlos Gonzalo; School of Engineering and Sciences; Campus MonterreyStress and depression are two major topics of concern for higher education institutions. Studies have shown how mental health problems can decrease students' ability to function efficiently during their education life and how depression can risk their physical well-being. To aid students in coping with the challenging experience of higher education and therefore enable them to perform better in stressful situations post-graduation, researchers recommend increasing their level of resilience. In an attempt to measure a person's resilience, previous studies have developed and analyzed self-rating questionnaires. While these studies have provided a way to assess people's psychological responses and provided a significant amount of insight, they do not provide an objective measurement for resilience to mental stress. There have been related studies that have evaluated physiological signals in individuals and have identified relationships with people's stress. Based on previous literature and applying machine learning, this thesis aims to demonstrate the feasibility of measuring an individual's resilience to mental stress and proposes a Resilience to Mental Stress Index (RMSI). In addition to this, this thesis presents an experiment's methodology to collect physiological and psychological data using smartwatch embedded sensors and psychological tools to study depression prediction. This research performed data analysis of 71 individuals subjected to a 10-minute psychophysiological test to study resilience to mental stress. The data collected considers five physiological features: (a) muscle response (electromyography), (b) blood volume pulse, (c) breathing rate, (d) peripheral temperature, and (e)skin conductance. We utilized unsupervised learning techniques to visualize and identify the relationship between these feature variability. As a result of the analysis, we created three different methods for the RMSI. The results' analysis between the different methods showed no statistically significant difference (p>0.05). However, we recommend using the Mahalanobis distance (MD) method because of its relationship with the validation methods. Even though there exists no standard method to quantify resilience to mental stress, our results indicate a positive relationship to the Resilience in Mexicans (RESI-M) psychological tool. For the study of depression prediction, during this research, five variables were selected for the study: (a) personality traits, (b) RMSI, (c) heart rate variability (HRV), and (d) sleeping disorders. To collect these variables, we developed a methodological framework and built a mobile application. We hope that this research serves as a solid baseline to understand resilience to mental stress and collect valuable information to predict depression.