Ettore Ritacco

The DAEDALUS framework: progressive querying and mining of movement data

In this work we propose DAEDALUS, a formal framework and system, specifically focussed on progressive combination of mining and querying operators. The core component of DAEDALUS is the MO-DMQL query language that extends SQL in two respects, namely a pattern definition operator and the capability to uniform manipulating both raw data and unveiled patterns. DAEDALUS system is specifically focussed on movement data and has been implemented as a query execution layer on top of the Hermes Moving Object Database. The expressiveness and usefulness of the MODMQL language as well as the computational capabilities of DAEDALUS are qualitatively evaluated by means of a case study.

X-Class: Associative Classification of XML Documents by Structure

The supervised classification of XML documents by structure involves learning predictive models in which certain structural regularities discriminate the individual document classes. Hitherto, research has focused on the adoption of prespecified substructures. This is detrimental for classification effectiveness, since the a priori chosen substructures may not accord with the structural properties of the XML documents. Therein, an unexplored question is how to choose the type of structural regularity that best adapts to the structures of the available XML documents. We tackle this problem through X-Class, an approach that handles all types of tree-like substructures and allows for choosing the most discriminatory one. Algorithms are designed to learn compact rule-based classifiers in which the chosen substructures discriminate the classes of XML documents. X-Class is studied across various domains and types of substructures. Its classification performance is compared against several rule-based and SVM-based competitors. Empirical evidence reveals that the classifiers induced by X-Class are compact, scalable, and at least as effective as the established competitors. In particular, certain substructures allow the induction of very compact classifiers that generally outperform the rule-based competitors in terms of effectiveness over all chosen corpora of XML data. Furthermore, such classifiers are substantially as effective as the SVM-based competitor, with the additional advantage of a high-degree of interpretability.

Effective XML Classification Using Content and Structural Information via Rule Learning

We propose a new approach to XML classification, that uses a particular rule-learning technique for the induction of interpretable classification models. These separate the individual classes of XML documents by looking at the presence within the XML documents themselves of certain features, that provide information on their content and structure. The devised approach induces classifiers with outperforming effectiveness in comparison to several established competitors.

Probabilistic Approaches to Recommendations

The importance of accurate recommender systems has been widely recognized by academia and industry, and recommendation is rapidly becoming one of the most successful applications of data mining and machine learning. Understanding and predicting the choices and preferences of users is a challenging task: real-world scenarios involve users behaving in complex situations, where prior beliefs, specific tendencies, and reciprocal influences jointly contribute to determining the preferences of users toward huge amounts of information, services, and products. Probabilistic modeling represents a robust formal mathematical framework to model these assumptions and study their effects in the recommendation process. This book starts with a brief summary of the recommendation problem and its challenges and a review of some widely used techniques Next, we introduce and discuss probabilistic approaches for modeling preference data. We focus our attention on methods based on latent factors, such as mixture models, probabilistic matrix factorization, and topic models, for explicit and implicit preference data. These methods represent a significant advance in the research and technology of recommendation. The resulting models allow us to identify complex patterns in preference data, which can be exploited to predict future purchases effectively. The extreme sparsity of preference data poses serious challenges to the modeling of user preferences, especially in the cases where few observations are available. Bayesian inference techniques elegantly address the need for regularization, and their integration with latent factor modeling helps to boost the performances of the basic techniques. We summarize the strengths and weakness of several approaches by considering two different but related evaluation perspectives, namely, rating prediction and recommendation accuracy. Furthermore, we describe how probabilistic methods based on latent factors enable the exploitation of preference patterns in novel applications beyond rating prediction or recommendation accuracy. We finally discuss the application of probabilistic techniques in two additional scenarios, characterized by the availability of side information besides preference data. In summary, the book categorizes the myriad probabilistic approaches to recommendations and provides guidelines for their adoption in real-world situations.

Fault detection and explanation through big data analysis on sensor streams

Abstract Fault prediction is an important topic for the industry as, by providing effective methods for predictive maintenance, allows companies to perform important time and cost savings. In this paper we describe an application developed to predict and explain door failures on metro trains. To this end, the aim was twofold: first, devising prediction techniques capable of early detecting door failures from diagnostic data; second, describing failures in terms of properties distinguishing them from normal behavior. Data pre-processing was a complex task aimed at overcoming a number of issues with the dataset, like size, sparsity, bias, burst effect and trust. Since failure premonitory signals did not share common patterns, but were only characterized as non-normal device signals, fault prediction was performed by using outlier detection. Fault explanation was finally achieved by exhibiting device features showing abnormal values. An experimental evaluation was performed to assess the quality of the proposed approach. Results show that high-degree outliers are effective indicators of incipient failures. Also, explanation in terms of abnormal feature values (responsible for outlierness) seems to be quite expressive.There are some aspects in the proposed approach that deserve particular attention. We introduce a general framework for the failure detection problem based on an abstract model of diagnostic data, along with a formal problem statement. They both provide the basis for the definition of an effective data pre-processing technique where the behavior of a device, in a given time frame, is summarized through a number of suitable statistics. This approach strongly mitigates the issues related to data errors/noise, thus enabling to perform an effective outlier detection. All this, in our view, provides the grounds of a general methodology for advanced prognostic systems.

From global to local and viceversa: uses of associative rule learning for classification in imprecise environments

We propose two models for improving the performance of rule-based classification under unbalanced and highly imprecise domains. Both models are probabilistic frameworks aimed to boost the performance of basic rule-based classifiers. The first model implements a global-to-local scheme, where the response of a global rule-based classifier is refined by performing a probabilistic analysis of the coverage of its rules. In particular, the coverage of the individual rules is used to learn local probabilistic models, which ultimately refine the predictions from the corresponding rules of the global classifier. The second model implements a dual local-to-global strategy, in which single classification rules are combined within an exponential probabilistic model in order to boost the overall performance as a side effect of mutual influence. Several variants of the basic ideas are studied, and their performances are thoroughly evaluated and compared with state-of-the-art algorithms on standard benchmark datasets.

Rule Learning with Probabilistic Smoothing

A hierarchical classification framework is proposed for discriminating rare classes in imprecise domains, characterized by rarity (of both classes and cases), noise and low class separability. The devised framework couples the rules of a rule-based classifier with as many local probabilistic generative models. These are trained over the coverage of the corresponding rules to better catch those globally rare cases/classes that become less rare in the coverage. Two novel schemes for tightly integrating rule-based and probabilistic classification are introduced, that classify unlabeled cases by considering multiple classifier rules as well as their local probabilistic counterparts. An intensive evaluation shows that the proposed framework is competitive and often superior in accuracy w.r.t. established competitors, while overcoming them in dealing with rare classes.

Survival Factorization on Diffusion Networks

In this paper we propose a survival factorization framework that models information cascades by tying together social influence patterns, topical structure and temporal dynamics. This is achieved through the introduction of a latent space which encodes: (a) the relevance of a information cascade on a topic; (b) the topical authoritativeness and the susceptibility of each individual involved in the information cascade, and (c) temporal topical patterns. By exploiting the cumulative properties of the survival function and of the likelihood of the model on a given adoption log, which records the observed activation times of users and side-information for each cascade, we show that the inference phase is linear in the number of users and in the number of adoptions. The evaluation on both synthetic and real-world data shows the effectiveness of the model in detecting the interplay between topics and social influence patterns, which ultimately provides high accuracy in predicting users activation times. Code and data related to this chapter are available at: https://doi.org/10.6084/m9.figshare.5411341.

Learning Effective XML Classifiers Based on Discriminatory Structures and Nested Content

Supervised classification aims to learn a model (or a classifier) from a collection of XML documents individually marked with one of a predefined set of class labels. The learnt classifier isolates each class by the content and structural regularities observed within the respective labeled XML documents and, thus, allows to predict the unknown class of unlabeled XML documents by looking at their content and structural features. The classification of unlabeled XML documents into the predefined classes is a valuable support for more effective and efficient XML search, retrieval and filtering.

Predicting Temporal Activation Patterns via Recurrent Neural Networks

We tackle the problem of predict whether a target user (or group of users) will be active within an event stream before a time horizon. Our solution, called PATH, leverages recurrent neural networks to learn an embedding of the past events. The embedding allows to capture influence and susceptibility between users and places closer (the representation of) users that frequently get active in different event streams within a small time interval. We conduct an experimental evaluation on real world data and compare our approach with related work.