Aniello De Santo

Content-Based Multimedia Retrieval

Content-based multimedia information retrieval (IR) provides new models and methods for effectively and efficiently “searching” through the huge variety of media that are available in different kinds of repositories (digital libraries, Web portals, social networks, multimedia databases, etc.). In this chapter, we will review the current state of the art of content-based multimedia information retrieval, including the most promising browsing and search paradigms for the several types of multimedia data, and show some cultural heritage applications.

SemTree: An index for supporting semantic retrieval of documents

In this paper, we propose SemTree, a novel semantic index for supporting retrieval of information from huge amount of document collections, assuming that semantics of a document can be effectively expressed by a set of 〈subject, predicate, object〉 statements as in the RDF model. A distributed version of KD-Tree has been then adopted for providing a scalable solution to the document indexing, leveraging the mapping of triples in a vectorial space. We investigate the feasibility of our approach in a real case study, considering the problem of finding inconsistencies in documents related to software requirements and report some preliminary experimental results.

Sentiment Detection in Social Networks Using Semantic Analysis: A Tool for Sentiment Analysis and Its Application in Cultural Heritage Realm

The spread of social networks as Twitter, Facebook, or Google+ or specialized ones as LinkedIn or Viadeo allows sharing opinions on different aspects of life every day. Millions of messages appear daily on the web thanks to blogs, microblogs, social networks, or review collector sites. This textual information can be divided in two main categories: facts and opinions. Facts are objective statements, while opinions reject and reveal people’s sentiments about products, personalities, and events and are extremely important when someone needs opinions before taking a decision. This information is a rich source of data for opinion mining. The interest that potential customers show in online opinions and reviews about products is something that vendors are gradually paying more and more attention to. In this scenario, a promising approach is sentiment analysis: the computational study of opinions, sentiments, and emotions expressed in a text. Its main aim is the identification of the agreement or disagreement statements that deal with positive or negative feelings in comments or reviews. In this chapter, we investigate the literature’s state of the art and the adoption of a probabilistic approach based on the latent Dirichlet allocation (LDA). By this approach, for a set of documents belonging to a same knowledge domain, a graph, the mixed graph of terms (mGTs), can be automatically extracted. This graph contains a set of weighted word pairs, which are discriminative for sentiment classification. The proposed method has been applied for the real-time analysis of documents, as TripAdvisor’s posts, in Italian language of opinion holders or social groups in the Databenc context: urban spaces, museums, archaeological parks, and events.

Commentary: Developmental Constraints on Learning Artificial Grammars with Fixed, Flexible, and Free Word Order

A long standing hypothesis in linguistics is that typological generalizations can shed light on the nature of the cognitive constraints underlying language processing and acquisition. In this perspective, Nowak and Baggio (2017) address the question of whether human learning mechanisms are constrained in ways that reflect typologically attested (possible) or unattested (impossible) linguistic patterns (Moro et al., 2001; Moro, 2016). Here, I show that the contrasts in Nowak and Baggio (2017) can be explained by language-theoretical characterizations of the stimuli, in line with a relatively recent research program focused on studying phonological generalizations from a mathematical perspective (Heinz, 2011a,b). The fundamental insight is that linguistic regularities that fall outside of certain complexity classes cannot be learned, due to computational properties reflecting implicit cognitive biases.

A Single Movement Normal Form for Minimalist Grammars

Movement is the locus of power in Minimalist grammars MGs but also their primary source of complexity. In order to simplify future analysis of the formalism, we prove that every MG can be converted into a strongly equivalent MG where every phrase moves at most once. The translation procedure is implemented via a deterministic linear tree transduction on the derivation tree language and induces at most a linear blow-up in the size of the lexicon.