GUIDED GENETIC PROGRAMMING MODELS FOR MULTIPLE INSTANCE LEARNING

Ph.D. Student: Amelia Zafra
Advisors: Sebastián Ventura
Abstract: This work focuses on the design of grammatical genetic programming models for solving different paradigm of learning applications with multiple instances.

DATA MINING FOR THE IMPROVEMENT OF ELEARNING COURSES

Ph.D. Student: Enrique García
Advisors: Cristóbal Romero, Carlos de Castro
Abstract: This thesis proposes a system oriented to find, share and suggest the most appropriate modifications to improve the effectiveness of the course. We use an iterative methodology to develop and carry out the maintenance of web-based courses to which we have added a specific data mining step.

HYBRID AND CONSTRUCTIVE METAHEURISTICS: METHODS AND APPLICATIONS

Ph.D. Student: Francisco Javier Rodríguez
Advisors: Carlos García, Manuel Lozano
Abstract: The main objective of this thesis is the development of optimisation methods for black-box environments based on hybrid metaheuristics combining simulated annealing and evolutionary algorithms, as well as the application of hybrid and constructive metaheuristics for non-uniform parallel machines scheduling and quadratic minimum spanning tree problems.

HYBRID MODEL FOR THE RECOMMENDATION OF LEARNING OBJECTS

Ph.D. Student: Alfredo Zapata
Advisors: Cristóbal Romero, Manuel Emilio Prieto
Abstract: This thesis proposes a collaborative methodology for searching, selecting and rating LOs in a group. We have implemented this methodology into a hybrid recommendation system called DELPHOS, which is a framework to assist users in the single/individual personalised search for learning objects in repositories.

ANT PROGRAMMING FOR CLASSIFICATION RULE MINING. APPLICATIONS

Ph.D. Student: Juan Luis Olmo
Advisors: Sebastián Ventura, José Raúl Romero
Abstract:The main objective of this thesis is the development of automatic programming models based on ant colony optimization toh2 address classification problems.

GENETIC PROGRAMMING FOR MULTI-LABEL CLASSIFICATION

Ph.D. Student: José Luis Ávila
Advisors: Sebastián Ventura, Eva L. Gibaja
Abstract: The main objective of this thesis is the development of a series of genetic programming models to solve multilabel classification problems. To do it, we propose the development of a multi-label classification using a genetic programming approach that allows working with numerical, categorical and nominal data sets.

NEW CHALLENGES IN ASSOCIATION RULE MINING: AN APPROACH BASED ON GENETIC PROGRAMMING

Ph.D. Student: José María Luna
Advisors: Sebastián Ventura, José Raúl Romero, Cristóbal Romero
Abstract: This thesis involves a series of approaches for mining association rules by means of a grammar-guided genetic programming based methodology. The ultimate goal is to provide new algorithms that mine association rules in only one step and in a highly efficient way. The use of grammars enables the flexibility of the extracted knowledge to be increased. Grammars also enable obtaining association rules that comprise categorical, quantitative, positive and negative attributes to be mined.

NEW CLASSIFICATION MODELS THROUGH EVOLUTIONARY ALGORITHMS

Ph.D. Student: Alberto Cano
Advisors: Sebastián Ventura, Amelia Zafra
Abstract: The objective of this thesis is the development of classification models using evolutionary algorithms, focusing on the aspects of scalability, interpretability and accuracy in complex datasets and high dimensionality.

PREDICTING STUDENT FAILURE AND DROPOUT AT SCHOOL USING DATA MINING TECHNIQUES

Ph.D. Student: Carlos Márquez
Advisors: Sebastián Ventura, Cristóbal Romero
Abstract: This study proposes to predict school failure in secondary education by using DM and to detect the factors that most influence school failure in young students by using classification techniques.

NEW HYBRID LEARNING MODELS FOR MULTI-LABEL CLASSIFICATION AND LABEL RANKING

Ph.D. Student: Oscar Gabriel Reyes
Advisors: Sebastián Ventura
Abstract: This doctoral thesis is focused on the multi-label learning paradigm. Two problems were studied, firstly, improving the performance of the algorithms on complex multi-label data, and secondly, improving the performance through unlabeled data.

IMPROVING EDUCATIONAL PROCESS MINING DISCOVERING BY GROPING DATA ABOUT STUDENT’S INTERACTION WITH MOODLE

Ph.D. Student: Alejandro Bogarín
Advisors: Cristóbal Romero, Rebeca Cerezo
Abstract: In this thesis, we propose to use Educational Process Mining (EPM) techniques to discover, analyze and provide a visual representation of the complete educational process in order to make unexpressed knowledge explicit and to facilitate better understanding of the educational process.

METAHEURISTIC MODELS TO THE DEVELOPMENT OF DECISION SUPPORT SYSTEMS IN SOFTWARE CONSTRUCTION

Ph.D. Student: Aurora Ramírez
Advisors: José Raúl Romero, Sebastián Ventura
Abstract: With this thesis, we plan to explore the synergies between the artificial intelligence and software engineering in order to define and solve new problems within the field of search-based software engineering (SBSE), especially those focused on early software conception and analysis.

NOVEL SUPPORT VECTOR MACHINES FOR DIVERSE LEARNING PARADIGMS

Ph.D. Student: Gabriella Melki
Advisors: Alberto Cano, Sebastián Ventura
Abstract: In this thesis, three multi-target support vector regression (SVR) models are presented. Then, under the multi-instance paradigm, a novel SVM multiple-instance formulation and an algorithm with a bag-representative selector are presented. Finally, a novel stochastic, i.e. online, learning algorithm for solving the L1-SVM problem in the primal domain is presented.

DISTRIBUTED MULTI-LABEL LEARNING ON APACHE SPARK

Ph.D. Student: Jorge González
Advisors: Alberto Cano, Sebastián Ventura
Abstract: This thesis proposes a series of multi-label learning algorithms for classification and feature selection implemented on the Apache Spark distributed computing model. Five approaches for determining the optimal architecture to speed up the multi-label learning methods, three distributed multi-label k nearest neighbors methods built on top of the Spark architecture, and two distributed feature selection methods for multi-label problems are proposed.

NEW CHALLENGES ON ASSOCIATIVE CLASSIFICATION: BIG DATA AND APPLICATIONS

Ph.D. Student: Francisco Solano Padillo Ruz
Advisors: Jose M. Luna, Sebastián Ventura
Abstract: The main objective of this thesis is to solve the challenging problem of Associative Classification and its application on very large datasets.

TEXT MINING AND MEDICINE: AN APPROACH TO EARLY DISEASE DETECTION

Ph.D. Student: Mª del Carmen Luque
Advisors: Jose M. Luna, Sebastián Ventura
Abstract: The main objective of this thesis is to develop a system based on Text Mining capable of transforming the textual clinical information in Knowledge to support the health professional in making decisions that allow the early detection of a disease.

MULTI-LABEL CLASSIFICATION MODELS FOR HETEROGENEOUS DATA: AN ENSEMBLE-BASED APPROACH

Ph.D. Student: Jose María Moyano
Advisors: Eva L. Gibaja, Krzysztof J. Cios, Sebastián Ventura
Abstract: The main objective of this thesis is to develop supervised learning models based on ensembles for problems with more flexible data representation for both the input and output space, as well as their application to solve real problems.

PROPOSITIONALIZATION METHODS IN THE SOLUTION OF MULTI-RELATIONAL DATA MINING PROBLEMS

Ph.D. Student: Luis A. Quintero-Domínguez
Advisors: Carlos Morell, Sebastián Ventura
Abstract: The main goal of this thesis is the development of new propositionalization methods that reduce the information loss to improve the accuracy of the learning algorithms. It involves designing and developing a propositionalization method that takes advantage of the potentialities of multi-instance representation as well as a a propositionalization method based on grammar-guided genetic programming.

MODELING COURSE DIFFICULTY INDEXES TO ENHANCE STUDENTS PERFORMANCE AND COURSE STUDY PLANS

Ph.D. Student: Mohammed Al-Twijri
Advisors: Sebastián Ventura, Francisco Herrera
Abstract: This thesis addresses Long-Term Course Planning by proposing a sequential pattern mining approach to analyze study plans, a course difficulty index for eligibility assessment, and a web application for obtaining difficulty indices. The proposals are applied to real data from King Abdulaziz University. The findings emphasize the importance of considering factors beyond credit hours in curriculum design, such as course difficulty indices and proper activity sequencing in order to an effective for course selection.

AUTOMATIC DIAGNOSIS OF MELANOMA WITH MODERN MACHINE LEARNING TECHNIQUES

Ph.D. Student: Eduardo Pérez
Advisors: Sebastián Ventura
Abstract: The main goal of this thesis is the development of modern machine learning methods for the automatic (or semi-automatic) diagnosis of melanoma at early stages. This type of skin cancer has an increasing incidence in white people, causing close to 90% of skin cancer mortality.

DEVELOPMENT AND APPLICATION OF NEW MACHINE LEARNING MODELS FOR THE COLORECTAL CANCER STUDY

Ph.D. Student: Jose Antonio Delgado
Advisors: Carlos García, Sebastián Ventura
Abstract: The main objective of this thesis is to deploy machine learning models that focus not only on predicting the survival of colorrectal cancer patients but also on predicting the factors that could produce some complications in those patients.

CLASSIFICATION TECHNIQUES FOR AIRFARE FORECASTING

Ph.D. Student: Marco Antonio Barón
Advisors: Jose M. Luna, Sebastián Ventura
Abstract: This work focuses on addressing multi-factorial problems faced by commercial airlines, specifically the pricing war and dynamic discount table creation. The thesis proposes a methodology using evolutionary algorithms and data mining methods to predict, analyze, and interpret airline fares, mimicking manual processes carried out by pricing teams. It also introduces a novel methodology for customer acquisition utilizing a grammatically evolutionary feature selection algorithm.

APPLICATION OF SCIENTIFIC WORKFLOWS TO DATA-INTENSIVE DOMAINS

Ph.D. Student: Rubén Salado
Advisors: José Raúl Romero
Abstract: The aim of this Ph.D. thesis is to bring the processing of large amounts of data closer to scientists, independently of their computation skills. To this end, it is necessary to develop new mechanisms that facilitate the use of data science techniques and technologies to computing non-experts.

DEMOCRATIZATION OF ADVANCED MODELS FOR DATA SCIENCE

Ph.D. Student: Rafael Barbudo
Advisors: José Raúl Romero, Aurora Ramírez
Abstract: The main objective of this thesis is to develop AutoML approaches that assist both data scientist and end users during the knowledge extraction process.

EXTRACTING USEFUL KNOWLEDGE IN BIOMEDICINE THROUGH PATTERN MINING TECHNIQUES

Ph.D. Student: Antonio Manuel Trasierras
Advisors: Jose M. Luna, Sebastián Ventura
Abstract: With this thesis we aim to use classic pattern mining techniques and new techniques such as supervised descriptive pattern mining, in order to study and extract useful knowledge from genomic data without previous hypotheses nor apriori knowledge.

IMPROVING THE ARCHITECTURE AND INTERPRETABILITY OF DEEP NEURAL NETWORKS: AN EVOLUTIONARY COMPUTING APPROACH

Ph.D. Student: Antonio R. Moya
Advisors: Sebastián Ventura
Abstract: The aim of this thesis is to use evolutionary algorithms to optimise architectures and hyper-parameters on deep learning models. In order to reduce the high cost of these optimisation techniques, the search for these evolutionary algorithms is guided by the use of probabilistic models that predict the goodness of each model based on a partial evaluation.

NEW MODELS FOR ANOMALY DETECTION IN DATA STREAMS

Ph.D. Student: Aurora Esteban
Advisors: Amelia Zafra, Sebastián Ventura
Abstract: The main objective of this thesis is to advance in the development of models for anomaly detection in data stream applied to different domains such us predictive maintenance and with special emphasis in interpretable and explanatory results.

NEW DEEP LEARNING APPROACHES IN ANOMALY DETECTION. APPLICATIONS

Ph.D. Student: Mohammed Yahya
Advisors: Sebastián Ventura
Abstract: The main goal of the thesis is the development of new anomaly detection methods and their application to different real problems. The primary interest is the development of deep learning methods for that purpose, while the second objective of this research will be the validation of the proposed models to solve a series of real-world problems.

APPLICATION OF MACHINE LEARNING TECHNIQUES FOR THE AUTOMATION OF SYSTEMATIC REVIEW OF SCIENTIFIC LITERATURE

Ph.D. Student: José de la Torre-López
Advisors: José Raúl Romero, Aurora Ramírez
Abstract: This thesis proposal aims to explore the automation of systematic literature reviews (SLRs) using artificial intelligence techniques. The main objective of this thesis is integrating machine learning to automate certain phases of the SLR process while allowing human intervention for verification and refinement.

GRAPH REPRESENTATION BY USING MACHINE LEARNING

Ph.D. Student: Ghaidaa Ahmed
Advisors: Sebastián Ventura
Abstract: The objectives of this thesis are the development of new graph machine learning methods, and their validation to solve a series of real-world problems.

DEEP LEARNING MODELS WITH MULTIPLE INSTANCE DATA

Ph.D. Student: Mustafa A. Jalil
Advisors: Sebastián Ventura
Abstract: The main goal of this research is the development of new multiple-instance models based on deep networks, and their application to different real problems.

IMPROVING PREDICTIVE ANALYSIS WITH LONGITUDINAL DATA USING DEEP LEARNING MODELS

Ph.D. Student: Roula Kadhim
Advisors: Sebastián Ventura
Abstract: The main goal of this thesis is to improve longitudinal data modelling by using deep learning methods and their applications to solve many different real-world applications in different sensitive fields.

NEW CHALLENGES IN DEVELOPMENT OF REUSABLE MACHINE LEARNING MODELS

Ph.D. Student: Zaid Zubair
Advisors: Sebastián Ventura
Abstract: The main objective of this thesis is the development and implementation of new Lifelong Machine Learning algorithms to use the previously learned information in future learning in an efficient way to reduce the exhausted of memory and time of learning, while overcoming the catastrophic forgetting problem.

IMPROVING THE EXPLOITATION OF RADIOMIC DATA USING ARTIFICIAL INTELLIGENCE TECHNIQUES

Ph.D. Student: Eduardo Almeda
Advisors: Sebastián Ventura, Jose M. Luna
Abstract: The main objective of this thesis is the study of features extracted by radiomics in medical images, improving the performance of the methodology, increasing the interpretability of the solutions and applying descriptive analysis techniques to improve the understanding of the images.

NEW MODELS FOR DATA AUGMENTATION IN TIME SERIES PROBLEMS

Ph.D. Student: Álvaro Espejo
Advisors: José Luis Ávila, Sebastián Ventura
Abstract: The objective of thesis to devise methods that allow the generation of synthetic time series data samples using generative methods in order to work in a machine learning environment with few labelled or unbalanced data sets.

DATA AUGMENTATION TECHNIQUES FOR BIOSIGNAL ANOMALY DETECTION.

Ph.D. Student: Mohammed Ayoub
Advisors: Sebastián Ventura
Abstract: The objective of this study is to investigate various data augmentation techniques in time series anomaly detection. These techniques will be applied on biosignals such as EEG data to diagnosis, for example, the risk of developing Central Neuropathic Pain or other related neurological disorders.

COUNTERFACTUAL EXPLANAIBLE METHODS AND THEIR APPLICATION TO PREDICTIVE PROBLEMS IN EDUCATION

Ph.D. Student: Ilias Naser
Advisors: José Raúl Romero, Aurora Ramírez
Abstract: The main objetive of this tesis is to design and implement new methods that generate counterfactual explanations, allowing users to adapt them according to their needs. The goal is to improve existing methods, conduct comparative studies, and evaluate the quality of explanations from the user’s perspective, with a focus on applying these methods in educational data mining to enhance predictions and facilitate targeted interventions for at-risk students.