Logical Analysis of Data as a tool for the analysis of Probabilistic Discrete Choice Behavior

Abstract

Probabilistic Discrete Choice Models (PDCM) have been extensively used to interpret the behavior of heterogeneous decision makers that face discrete alternatives. The classification approach of Logical Analysis of Data (LAD) uses discrete optimization to generate patterns, which are logic formulas characterizing the different classes. Patterns can be seen as rules explaining the phenomenon under analysis. In this work we discuss how LAD can be used as the first phase of the specification of PDCM. Since in this task the number of patterns generated may be extremely large, and many of them may be nearly equivalent, additional processing is necessary to obtain practically meaningful information. Hence, we propose computationally viable techniques to obtain small sets of patterns that constitute meaningful representations of the phenomenon and allow to discover significant associations between subsets of explanatory variables and the output. We consider the complex socioeconomic problem of the analysis of the utilization of the Internet in Italy, using real data gathered by the Italian National Institute of Statistics.

Imputation techniques for the reconstruction of missing interconnected data from higher Educational Institutions

“…Hence, it could be adapted to impute different datasets containing educational data from other origins, or even datasets with different meaning but sharing the feature of interconnection among the variables. Future work includes the integration in the proposed methodology with the web scraping techniques described in [34,35] by using the Universities' websites, or with the Logic-based techniques described in [36,37] to extract datasupported logic descriptions of the Institutions.…”

Identifying e-Commerce in Enterprises by means of Text Mining and Classification Algorithms

“…In particular, SVM techniques are generally able to give more stable results, since they are less sensible to overfitting issues, and they also exhibit a better flexibility with respect to the other classifiers (and for these reasons SVMs are often used in practical applications; see, e.g., [24][25][26]). SLAD, on the other hand, offers the advantages of providing intelligible logic rules for the classifications, which could be used for gaining more insight on the analyzed phenomenon (see, e.g., [27]). …”

Exploring the Potentialities of Automatic Extraction of University Webometric Information

“…This is a prominent field of research due to the continuous expansion of the Internet and the consequent demand for always more effective information retrieval strategies. Recent web mining approaches take advantage from the integration of natural language processing with many advanced machine learning techniques, such as classification algorithms (Bruni & Bianchi, 2015) or logic-based information processing tools (Bruni et al, 2019).…”