Using knowledge-based relatedness for information retrieval

What’s New? Analysing Language-Specific Wikipedia Entity Contexts to Support Entity-Centric News Retrieval

“…Examples include information retrieval [4,14,18], named entity disambiguation [1,2,7,8,11,12], text classification [25] and entity ranking [10]. To extract the content of an entity context, many researches directly used the Wikipedia article describing the entity [1,2,8,9,14,[25][26][27]; some works extended the article with all the other Wikipedia articles linked to the Wikipedia article describing the entity [6,7,12]; while some only considered the first paragraph of the Wikipedia article describing the entity [2].…”

“…As for incorporating the Wikipedia knowledge in information retrieval applications, [4,15,18] applied concept-based approaches that mapped both the documents and queries to the Wikipedia concept space; [14,23] focused only on query extension; [20,24] focused only on mapping documents to Wikipedia concept space. To retrieve documents that did not explicitly mention the query entity by name, but were still relevant to the query entity, we chose to map both the query and the documents to the aspect space.…”

Automatic Wordnet Development for Low-Resource Languages using Cross-Lingual WSD

“…Instead, new wordnets could be developed by adopting the structure of existing wordnets in other languages (usually WordNet) and translating the words associated with their synsets into the target language. One important advantage of this approach is that the resulting wordnet is aligned to the WordNet and the ILI, and thus is interesting for contrastive semantic analysis and is particularly useful in multi-lingual tasks such as multi-lingual information retrieval (Dini, Peters, Liebwald, Schweighofer, Mommers, & Voermans, 2005;Otegi, Arregi, Ansa, & Agirre, 2015) and multi-lingual semantic web (Buitelaar & Cimiano, 2014). The main assumption on which one can develop a wordnet using the expansion approach is that most of the concepts and semantic relations are common among different languages.…”

From senses to texts: An all-in-one graph-based approach for measuring semantic similarity

“…Measuring the semantic similarity of text pairs enables the evaluation of the output quality of machine translation systems [1] or the recognition of paraphrases [2], while laying the foundations for other fields, such as textual entailment [3,4], information retrieval [5,6], question answering [7,8], and text summarization [9]. At the word level, semantic similarity can have direct benefits for areas such as lexical substitution [10] or simplification [11], and query expansion [12], whereas, at the sense level, the measurement of semantic similarity of concept pairs can be utilized as a core component in many other applications, such as reducing the granularity of lexicons [13,14], Word Sense Disambiguation [15], knowledge enrichment [16], or alignment and integration of different lexical resources [17][18][19][20].…”