Sebastián Maldonado

University of Chile · Departamento de Control de Gestión y Sistemas de Información

In this paper, we introduce a novel predict-and-optimize method for profit-driven churn prevention. We frame the task of targeting customers for a retention campaign as a regret minimization problem. The main objective is to leverage individual customer lifetime values (CLVs) to ensure that only the most valuable customers are targeted. In contrast...

In this paper, we propose a fuzzy adaptive loss function for enhancing deep learning performance in classification tasks. Specifically, we redefine the cross-entropy loss to effectively address class-level noise conditions, including the challenging problem of class imbalance. Our approach introduces aggregation operators, leveraging the power of f...

As in any other task, the process of building machine learning models can benefit from prior experience. Meta-learning for classifier selection gains knowledge from characteristics of different datasets and/or previous performance of machine learning techniques to make better decisions for the current modeling process. Meta-learning approaches firs...

In this paper, we present a novel method that extends the kernel-based support vector regression to hierarchical time series analysis. This predictive task consists of taking advantage of the hierarchical structure of a set of related time series. This is a common challenge in retail, for example, in which product sales are grouped according to cat...

MALDI–TOF mass spectrometry (Matrix-Assisted Laser Desorption-Ionization (MALDI) and a Time-of-Flight detector (TOF) is a promising strategy for identifying patterns in data, establishing a relevant methodology for rapid and accurate microorganisms identification. However, this type of data is challenging to analyze due to its high complexity, and...

Dataset shift is a relevant topic in unsupervised learning since many applications face evolving environments, causing an important loss of generalization and performance. Most techniques that deal with this issue are designed for data stream clustering, whose goal is to process sequences of data efficiently under Big Data. In this study, we claim...

We present a comprehensive case study to integrate students into several concepts related to integer linear programming. The case article starts with a relatively simple scheduling/assignment problem. Then, the problem incorporates new elements to present various modeling principles incrementally. Each variation of the case enables the instructor t...

Funding: This work was supported by the Fondo Nacional de Desarrollo Científico y Tecnológico [Grant 1200221] and the Agencia Nacional de Investigación y Desarrollo [Grant AFB180003].

It is of utmost importance for marketing academics and service industry practitioners to understand the factors that influence customer satisfaction. This study proposes a novel framework to analyze open-ended survey data and extract drivers of customer satisfaction. This is done automatically via deep learning models for natural language processin...

Debt collection is a very important business application of predictive analytics. This task consists of foreseeing repayment chances of late payers. In this sense, contact centers have a central role in debt collection since it improves profitability by turning monetary losses into a direct benefit to banks and other financial institutions. In this...

In this paper, we propose a novel machine learning approach based on robust optimization. Our proposal defines the task of maximizing the two class accuracies of a binary classification problem as a Cobb-Douglas function. This function is well known in production economics and is used to model the relationship between two or more inputs as well as...

Stated-choice experiments have been useful in helping to make a number of operations management decisions. Many recent advances in this area have raised questions about estimating consumers’ preferences when they partially ignore the information provided in discrete choice experiments, a problem introduced as attribute non-attendance (ANA). This li...

Model selection is a highly important step in the process of extracting knowledge from datasets. This is usually done via partitioning strategies such as cross-validation in which the training and test subsets are selected randomly. However, it has been suggested in the literature that this is not the best approach in changing environments due to t...

The Synthetic Minority Over-sampling Technique (SMOTE) is a well-known resampling strategy that has been successfully used for dealing with the class-imbalance problem, one of the most challenging pattern recognition tasks in the last two decades. In this work, we claim that SMOTE has an important issue when defining the neighborhood in order to cr...

In several applications, there are hierarchically-organized time series that can be aggregated at various levels. In this paper, a novel Support Vector Regression approach is proposed for dealing with hierarchical time series forecasting. The main idea is to pool information across levels of hierarchy, preventing bottom-level series from deviate mu...

In this paper, we present a novel approach for n-gram generation in text classification. The a-priori algorithm is adapted to prune word sequences by combining three feature selection techniques. Unlike the traditional two-step approach for text classification in which feature selection is performed after the n-gram construction process, our propos...

Over the last two decades, governments have increased their investment in information technology to improve the use of public resources, using public electronic procurement systems to obtain better prices, better solutions and to show transparency in the procurement process. Public procurement of software development projects is specific acquisitio...

The decoy effect reveals a potential violation of the regularity assumption, which is a building block of canonical discrete choice models. This effect has been detected in various choice contexts, but the susceptibility to it and its subjective valuation have been scarcely studied before. This paper proposes, illustrates and assesses two methodolo...

Traffic pumping is a type of fraud committed in several countries, in which small telephone operators inflate the number of incoming calls to their networks, profiting from a higher access charge in relation to the network operator associated with the origin of the call. The identification of traffic pumping is complex due to the lack of labels for...

The predictive performance of classification methods relies heavily on the nature of the environment, as in the joint distribution of inputs and outputs may evolve over time. This issue is known as dataset shift. Given that most statistical and machine learning techniques assume that the training sample is drawn from the same distribution as the te...

Student dropout is a major concern in higher education, as it leads to direct economic losses and substantial social costs. Public and private institutions spend considerable resources to prevent student dropout. The efficiency and effectiveness of these investments, however, may be improved by adopting a profit-driven perspective. In this paper, w...

Functional Data Analysis (FDA) has become a very important field in recent years due to its wide range of applications. However, there are several real-life applications in which hybrid functional data appear, i.e., data with functional and static covariates. The classification of such hybrid functional data is a challenging problem that can be han...

Our paper proposes a novel approach for profit-based classification for churn prediction in the mutual fund industry. The maximum profit measure is redefined to address multiple segments that differ strongly in the average customer lifetime values (CLVs). The proposed multithreshold framework for churn prediction aims to maximize the profit of rete...

Classification is a well-studied machine learning task which concerns the assignment of instances to a set of outcomes. Classification models support the optimization of managerial decision-making across a variety of operational business processes. For instance, customer churn prediction models are adopted to increase the efficiency of retention ca...

In this paper, we propose a novel support vector regression (SVR) approach for time series analysis. An efficient forward feature selection strategy has been designed for dealing with high-frequency time series with multiple seasonal periods. Inspired by the literature on feature selection for support vector classification, we designed a technique...

Uplift modeling is an approach for estimating the incremental effect of an action or treatment at the individual level. It has gained attention in the marketing and analytics communities due to its ability to adequately model the effect of direct marketing actions via predictive analytics. The main contribution of our study is the implementation of...

In this study, a methodology to develop hourly demand scenarios in a medium‐term horizon for primary distribution substations is presented and applied to a case study. The main contribution of this study is that it addresses successfully the effect of saturation of distribution feeders in the medium term due to sustained growth in demand. In additi...

Inter-Organizational Information Systems (IOISs) for seaport logistics facilitate monitoring operations, the exchange of information with stakeholders, and meeting regulations of foreign trade. However, seaport contexts entail complexities in terms of stakeholder involvement and business processes that must be considered thoroughly toward the succe...

In this work, a novel method called epsilon-nonparallel support vector regression (ε-NPSVR) is proposed. The reasoning behind the nonparallel support vector machine (NPSVM) method for binary classification is extended for predicting numerical outputs. Our proposal constructs two nonparallel hyperplanes in such a way that each one is closer to one o...

In this paper, we propose three novel profit-driven strategies for churn prediction. Our proposals extend the ideas of the Minimax Probability Machine, a robust optimization approach for binary classification that maximizes sensitivity and specificity using a probabilistic setting. We adapt this method and other variants to maximize the profit of a...

In this study, an expert system is presented for analyzing the mental workload of interacting with a mobile phone while facing common daily tasks. Psychophysiological signals were collected from various devices, each characterized by a different cost and obtrusiveness. To deal with user-level signal data, a support vector machine-based feature sele...

A weighting strategy for handling outliers in binary classification using Support Vector Machine (SVM) is proposed in this work. The traditional SVM model is modified by introducing an Induced Ordered Weighted Averaging (IOWA) operator, in which the hinge loss function becomes an ordered weighted sum of the SVM slack variables. These weights are de...

In this work, a strategy for automatic lag selection in time series analysis is proposed. The method extends the ideas of feature selection with support vector regression, a powerful machine learning tool that can identify nonlinear patterns effectively thanks to the introduction of a kernel function. The proposed approach follows a backward variab...

A novel binary classification approach is proposed in this paper, extending the ideas behind nonparallel support vector machine (NPSVM) to robust machine learning. NPSVM constructs two twin hyperplanes by solving two independent quadratic programming problems and generalizes the well-known twin support vector machine (TWSVM) method. Robustness is c...

A novel framework for profit-based credit scoring is proposed in this work. The approach is based on robust optimization, which is designed for dealing with uncertainty in the data, and therefore is effective at classifying new samples that follow a slightly different distribution in relation to the original dataset used to create the model. Instea...

Times series often offers a natural disaggregation in a hierarchical structure. For example, product sales can come from different cities, districts, or states; or be grouped by categories and subcategories. This hierarchical structure can be useful for improving the forecast, and this strategy is known as hierarchical time series (HTS) analysis. I...

A Decision Support System (DSS)is proposed in this paper for improving container stacking operations. This DSS addresses the stacking problem for import containers via a two-step strategy. First, dwell times are predicted for each container using analytics techniques. This prediction is used as an input for a mathematical programming model that min...

In this paper, we propose novel second-order cone programming formulations for binary classification, by extending the Minimax Probability Machine (MPM) approach. Inspired by Support Vector Machines, a regularization term is included in the MPM and Minimum Error Minimax Probability Machine (MEMPM) methods. This inclusion reduces the risk of obtaini...

In this work, the Synthetic Minority Over-sampling Technique (SMOTE) approach is adapted for high-dimensional binary settings. A novel distance metric is proposed for the computation of the neighborhood for each minority sample, which takes into account only a subset of the available attributes that are relevant for the task. Three variants for the...

Credit scoring is a crucial task within risk management for any company in the financial sector. On the one hand, it is in the self-interest of banks to avoid approving credits to customers who probably default. On the other hand, regulators require strict risk management systems from banks to protect their customers and, from “too big to fail inst...

In the current copper mining scenario, where prices are decreasing and pits are larger, there is a pressing need for increasing operational productivity. This is particularly important for mining contractors, who are constantly facing the additional pressure of obsolescence if they are not able to provide cost-savings for mine owners. In this paper...

In this work, a data mining framework is proposed to improve the understanding of how people allocate their time. Using a multivariate approach, we performed a clustering procedure, and subsequently a regression analysis to detect which variables influence individual time use for each cluster found. Results suggest that the impact of various sociod...

In this paper, we propose a novel method for Support Vector Regression (SVR) based on second-order cones. The proposed approach defines a robust worst-case framework for the conditional densities of the input data. Linear and kernel-based second-order cone programming formulations for SVR are proposed, while the duality theory allows us to derive i...

Twin Support Vector Regression is an effective machine learning strategy, which splits the predictive task into two small problems, gaining in both efficiency and predictive performance. In this paper, a novel extension for twin Support Vector Regression is presented. The proposal is based on robust optimization, conferring robustness to the predic...

In this work, we propose a novel feature selection approach designed to deal with two major issues in machine learning, namely class-imbalance and high dimensionality. The proposed embedded strategy penalizes the cardinality of the feature set via the scaling factors technique, and is used with two support vector machine (SVM) formulations designed...

In this work, a novel nearest neighbor approach is presented. The main idea is to redefine the distance metric in order to include only a subset of relevant variables, assuming that they are of equal importance for the classification model. Three different distance measures are redefined: the traditional squared Euclidean, the Manhattan, and the Ch...

In this work, the classical soft-margin Support Vector Machine (SVM) formulation is redefined with the inclusion of an Ordered Weighted Averaging (OWA) operator. In particular, the hinge loss function is rewritten as a weighted sum of the slack variables to guarantee adequate model fit. The proposed two-step approach trains a soft-margin SVM first...

We have developed a new methodology for examining and extracting patterns from brain electric activity by using data mining and machine learning techniques. Data was collected from experiments focused on the study of cognitive processes that might evoke different specific strategies in the resolution of math problems. A binary classification proble...

All mathematical problems.

Multiclass classification is an important task in pattern analysis since numerous algorithms have been devised to predict nominal variables with multiple levels accurately. In this paper, a novel support vector machine method for twin multiclass classification is presented. The main contribution is the use of second-order cone programming as a robu...

“Data is the new oil” is just one of the sayings that describe the importance of data for today´s society. We have witnessed a rapid development of methods to analyze such data; starting with Statistics in the early 18th century, followed by Artificial Intelligence and Machine Learning, and finally leading to Data Science incorporating classical me...

In this work, two novel formulations for embedded feature selection are presented. A second-order cone programming approach for Support Vector Machines is extended by adding a second regularizer to encourage feature elimination. The one- and the zero-norm penalties are used in combination with the Tikhonov regularization under a robust setting desi...

n this paper, we propose a profit-driven approach for classifier construction and simultaneous variable selection based on linear Support Vector Machines. The main goal is to incorporate business-related information such as the variable acquisition costs, the Type I and II error costs, and the profit generated by correctly classified instances, int...

Recruiting prospective students efficiently and effectively is a very important challenge for universities, mainly because of the increasing competition and the relevance of enrollment-generated revenues. This work provides an intelligent system for modeling the student enrollment decisions problem. A nested logit classifier was constructed to pred...

Clustering is one of the main data mining tasks with many proven techniques and successful real-world applications. However, in changing environments, the existing systems need to be regularly updated in order to describe in the best possible way an observed phenomenon at each point in time. Since changes lead to uncertainty, the respective systems...

This paper presents a novel embedded feature selection approach for Support Vector Machines (SVM) in a choice-based conjoint context. We extend the L1-SVM formulation and adapt the RFE-SVM algorithm to conjoint analysis to encourage sparsity in consumer preferences. This sparsity can be attributed to consumers being selective about the attributes t...

Kernel methods are very important in pattern analysis due to their ability to capture nonlinear relationships in datasets. The best known kernel-based technique is Support Vector Machine (SVM), which can be used for several pattern recognition tasks, including multiclass classification. In this paper, we focus on maximum margin classifiers for nonl...

In this work, a novel feature selection method for twin Support Vector Machine (SVM) is presented. The main idea is to combine two regularizers, namely the Euclidean and infinite norm to perform twin classification and variable selection simultaneously. This latter task is performed in a coordinated fashion, enabling that the same attributes are se...

In this work we propose two formulations based on Support Vector Machines for simultaneous classification and feature selection that explicitly incorporate attribute acquisition costs. This is a challenging task for two main reasons: the estimation of the acquisition costs is not straightforward and may depend on multivariate factors, and the inter...

Support Vector Machines (SVMs) has been successfully used to identify individuals' preferences in conjoint analysis. One of the challenges of using SVMs in this context is to properly control for preference heterogeneity among individuals to construct robust partworths. In this work we present a new technique that obtains all individual utility fun...

Selecting the relevant factors in a particular domain is of utmost interest in the machine learning community. This paper concerns the feature selection process for twin support vector machine (TWSVM), a powerful classification method that constructs two nonparallel hyperplanes in order to define a classification rule. Besides the Euclidean norm, o...

Second-order cone programming (SOCP) formulations have received increasing attention as robust optimization schemes for Support Vector Machine (SVM) classification. These formulations study the worst-case setting for class-conditional densities, leading to potentially more effective classifiers in terms of performance compared to the standard SVM f...

The Santiago Fire Department (from here referred to as SFD) lacks a fleet management strategy since their vehicles remain allocated in fixed fire stations, while the presence of seasonal patterns suggests that the frequency of events changes according to their geographical distribution. This fact has led to inequitable service in terms of response...

Expert systems often rely heavily on the performance of binary classification methods. The need for accurate predictions in artificial intelligence has led to a plethora of novel approaches that aim at correctly predicting new instances based on nonlinear classifiers. In this context, Support Vector Machine (SVM) formulations via two nonparallel hy...

Feature selection is an important machine learning topic, especially in high dimensional applications, such as cancer prediction with microarray data. This work addresses the issue of high dimensionality of feature selection for linear and kernel-based Support Vector Machines (SVMs) considering second-order cone programming formulations. These form...

This paper presents novel second-order cone programming (SOCP) formulations that determine a linear multi-class predictor using Support Vector Machines (SVMs). We first extend the ideas of OvO (One-versus-One) and OvA (One-versus-All) SVM formulations to SOCP-SVM, providing two interesting alternatives to the standard SVM formulations. Additionally...

This work addresses the issue of high dimensionality for linear multiclass Support Vector Machines (SVMs) using second-order cone programming (SOCP) formulations. These formulations provide a robust and efficient framework for classification, while an adequate feature selection process may improve predictive performance. We extend the ideas of SOCP...

Credit scoring is an automated, objective and consistent tool which helps lenders to provide quick loan decisions. It can replace some of the more mechanical work done by experienced loan officers whose decisions are intuitive but potentially subject to bias. Prospective borrowers may have a strong motivation to fraudulently falsify one or more of...

The share of the services offered via the Internet by nowadays banking companies is quickly growing, making of the understanding of online customers one of the major concerns. Data mining tools have proven their efficiency in addressing this challenge by providing unsupervised quantitative techniques to identify those segments of customers with sim...

An empirical framework for customer churn prediction modeling is presented in this work. This task represents a very interesting business analytics challenge, given its highly class imbalanced nature, and the presence of noisy variables that adversely affect the prediction capabilities of classification models. In this work, two SVM-based technique...

In this work we present a novel maximum-margin approach for multi-class Support Vector Machines based on second-order cone programming. The proposed method consists of a single optimization model to construct all classification functions, in which the number of second-order cone constraints corresponds to the number of classes. This is a key differ...

Uno de los grandes desafíos de la Minería de Datos aplicada al Análisis de Negocios es la selección de atributos para un modelo de clasificación. La mayoría de las técnicas de selección de atributos se basan en criterios de validación estadística, perdiendo en muchos casos el objetivo del negocio en sí mismo, lo que no necesariamente lleva a modelo...

The geographical distribution of the population of the city of Santiago, Chile, has changed significantly in recent years. In spite of this fact, the location of the fire stations has remained unchanged. We propose a model for the optimal location of the fire stations and a fleet assignment for the Santiago Fire Department (SFD), aimed at maximisin...

Churn prediction is an important application of classification models that identify those customers most likely to attrite based on their respective characteristics described by e.g. socio-demographic and behavioral variables. Since nowadays more and more of such features are captured and stored in the respective computational systems, an appropria...

Archaeological sites composed only of surficial lithics are widespread in arid environments. Numerical dating of such sites is challenging, however, and even establishing a relative chronology can be daunting. One potentially helpful method for assigning relative chronologies is to use lithic weathering, on the assumption that the most weathered ar...

We present an unsupervised method that selects the most relevant features using an embedded strategy while maintaining the cluster structure found with the initial feature set. It is based on the idea of simultaneously minimizing the violation of the initial cluster structure and penalizing the use of features via scaling factors. As the base metho...

Dataset shift is present in almost all real-world applications, since most of them are constantly dealing with changing environments. Detecting fractures in datasets on time allows recalibrating the models before a significant decrease in the model’s performance is observed. Since small changes are normal in most applications and do not justify the...

Data mining techniques are widely used by researchers and companies in order to solve problems in a myriad of domains. While these techniques are being adopted and used in daily activities, new operational challenges are encountered concerning the steps following this adoption. In this paper, the problem of updating and improving an existing cluste...