David S. Batista

Where in the Wikipedia is that answer? the XLDB at the GikiCLEF 2009 task

We developed a new semantic question analyser for a custom prototype assembled for participating in GikiCLEF 2009, which processes grounded concepts derived from terms, and uses information extracted from knowledge bases to derive answers. We also evaluated a newly developed named-entity recognition module, based in Conditional Random Fields, and a new world geoontology, derived from Wikipedia, which is used in the geographic reasoning process.

Semi-Supervised Bootstrapping of Relationship Extractors with Distributional Semantics

Semi-supervised bootstrapping techniques for relationship extraction from text iteratively expand a set of initial seed relationships while limiting the semantic drift. We research bootstrapping for relationship extraction using word embeddings to find similar relationships. Experimental results show that relying on word embeddings achieves a better performance on the task of extracting four types of relationships from a collection of newswire documents when compared with a baseline using TFIDF to find similar relationships.

Geographic signatures for semantic retrieval

Geographic Information Retrieval (GIR) systems rely on the identification and disambiguation of place names in documents to determine the region about which they are relevant. The place names are mapped into geographic concepts and used to assign an encompassing concept (a scope) to each document. However, sometimes a single scope is too restrictive and insufficient for capturing the geographic semantics of a document. We propose as an alternative to abstract the geographic semantics of a document as a geographic signature, which is a list of maximally disambiguated geographic references found in a document. A signature can be used in multiple GIR applications, such as in building a geographic index for a document collection. We perform the disambiguation of the possible geographic meanings using semantic similarity measures.

POWER - Politics Ontology for Web Entity Retrieval

POWER is an ontology of political processes. It is designed for tracking politicians, political organisations and elections, both in mainstream and social media. In social media, these entities (particularly humans) are frequently named by emergent abbreviations, non-standardized acronyms, nicknames, metaphoric expressions and neologisms. Politicians are also frequently mentioned in texts by their roles in the political scene, which may change rapidly over time. This paper describes how POWER was designed for tracking such complex and dynamic setting, with the purpose of making it a key resource to analytics applications mining the media.

Toponym disambiguation using ontology-based semantic similarity

We propose a new heuristic for toponym sense disambiguation, to be used when mapping toponyms in text to ontology concepts, using techniques based on semantic similarity measures. We evaluated the proposed approach using a collection of Portuguese news articles from which the geographic entity names were extracted and then manually mapped to concepts in a geospatial ontology covering the territory of Portugal. The results suggest that using semantic similarity to disambiguate toponyms in text produces good results, in comparison with a baseline method.

Document Analytics through Entity Resolution

We present a prototype system for resolving named entities, mentioned in textual documents, into the corresponding Wikipedia entities. This prototype can aid in document analysis, by using the disambiguated references to provide useful information in context.

A Minwise Hashing Method for Addressing Relationship Extraction from Text

Relationship extraction concerns with the detection and classification of semantic relationships between entities mentioned in a collection of textual documents. This paper proposes a simple and on-line approach for addressing the automated extraction of semantic relations, based on the idea of nearest neighbor classification, and leveraging a minwise hashing method for measuring similarity between relationship instances. Experiments with three different datasets that are commonly used for benchmarking relationship extraction methods show promising results, both in terms of classification performance and scalability.

Tracking Politics with POWER.

Purpose – POWER is an ontology of political processes and entities. It is designed for tracking politicians, political organizations and elections, both in mainstream and social media. The aim of this paper is to propose a data model to describe political agents and their relations over time.Design/methodology/approach – The authors propose a data model to describe political agents (politicans, political instutions and political associations) and their relations over time. The model is formalized as an ontology using the RDF format and the population is performed in two steps. First, a bootstrap process loads data collected from authoritative sources. Then, the ontology is enriched with alternative media names extracted from the web.Findings – The ontology is published as a public resource following the guidelines of linked data and semantic web standards can be accessed via SPARQL endpoint.Originality/value – The authors have developed an ontology for the political domain tailored to aid in the tasks of ...