China
Africa
Explore chapters and articles related to this topic
On Emerging Use Cases and Techniques in Large Networked Data in Biomedical and Social Media Domain
Published in Yulei Wu, Fei Hu, Geyong Min, Albert Y. Zomaya, Big Data and Computational Intelligence in Networking, 2017
Vishrawas Gopalakrishnan, Aidong Zhang
In this chapter, we have discussed the specific application of graph theory and techniques for entity resolution and hypothesis generation. One of the major common themes across the methodologies mentioned earlier, is the use of individual token(s) as a node in the graph. However, it is also possible to consider a phrase as a node, for instance the title of a web page. Such modeling is frequently employed in tasks where one wants to depict relationships between concepts or entities, like in collective ER. For more example in other related application areas, we would like to refer the readers to these works [52-54], where the objective of the authors is that given an input text, the system should link tokens to corresponding entries in a given knowledge base. As an example, given the text “Paris is a beautiful city,” the system should identify that the word “Paris" refers to the city Paris and not an individual named Paris. It should then link this token to the appropriate entry in the knowledge base. This task is called entity linking and is an important task when attempting a semantic analysis. For instance in the papers [55, 56] and also the ones mentioned earlier in the paragraph, the authors provide various mechanisms based on graph topology and traversal techniques (e.g., random walk) to determine relatedness, similarity, and matching.
Mapping Twitter hate speech towards social and sexual minorities: a lexicon-based approach to semantic content analysis
Published in Behaviour & Information Technology, 2020
Vittorio Lingiardi, Nicola Carone, Giovanni Semeraro, Cataldo Musto, Marilisa D’Amico, Silvia Brena
Semantic tagging was used to identify (and filter out) ambiguous Tweets. A Tweet was considered ambiguous when it contained one or more terms in the lexicon but lacked a clear intolerant intent. As described above, Tweet t2 was an example of a Tweet characterised by this issue. The Semantic Tagger implemented entity linking algorithms to better identify the meaning and intent of the content extracted by the Social Extractor. Generally speaking, the goal of entity linking is to identify the entities mentioned in a piece of text. While a complete discussion of entity linking algorithms is beyond the scope of this paper (we suggest that readers who are interested in this topic refer to Derczynski et al. 2015), in simple terms, the entity linking process uses statistical approaches to map portions of the input text to one or more entities by exploiting large knowledge bases, such as Wikipedia.
EREL: an Entity Recognition and Linking algorithm
Published in Journal of Information and Telecommunication, 2018
Cuong Duc Nguyen, Trong Hai Duong
Recognizing entity mentions in a text and linking them to entities in a knowledge base are two fundamental tasks in text analysis. In the knowledge extraction pipeline, a Named Entity Recognition (NER) system is often used to recognize mentions of named entities in text and then an Entity Linking (EL) system is executed to link recognized mentions to entities in a knowledge base like Wikipedia (Rao, McNamee, & Dredze, 2013). Because NER systems focus on identifying named entities, such as people, organizations and locations, EL is often considered as only linking named entities and not able processing nominal entities (Moro, Raganato, & Navigli, 2014).