Article Structure
Abstract
Sentiment relevance detection problems occur when there is a sentiment expression in a text, and there is the question of whether or not the expression is related to a given entity or, more generally, to a given situation.
Introduction
Sentiment extraction by modern sentiment analysis (SA) systems is usually based on searching the input text for sentiment-bearing words and expressions, either general (language-wide) or domain-specific.
Related Work
The task of SA has drawn the attention of many researchers worldwide (Connor et al., 2010; Liu, 2012; Loughran and Mcdonald, 2010; Pang and Lee, 2004; Turney, 2002).
Entity Relevance
An instance of the sentiment relevance detection problem for a single entity consists of a text document, a sentiment expression within the document, and a target entity.
Relevance Algorithms
Each algorithm receives, as input, the text of the document, with labeled reference of the target entity and other entities of the same type.
Experiments
For the experiments, we use two manually-annotated corpora2, a financial corpus3 and a medical4 corpus.
Conclusion
The results are mostly intuitively understood and confirm the expectations.
Topics
entity types
Appears in 8 sentences as: entities types (1) entity type (2) entity types (5)
In Entities' Sentiment Relevance
- Another layer that we'd like to add concerns the interaction of different entity types during SA.
Page 1, “Introduction”
- In a typical situation, there is only one entity type which is the target for SA.
Page 1, “Introduction”
- In such cases, clearly distinguishing between the relevancy of target and non-target entities types is not essential.
Page 1, “Introduction”
- We will show that such situations are modeled well enough using intersections of regions of relevance of the participating entity types , with the relevance region for each type calculated separately.
Page 1, “Introduction”
- The examples we are interested in are in the medical domain and deal with three main entity types : PERSON, DRUG, and DISEASE, where PERSON is restricted to known physicians.
Page 2, “Entity Relevance”
- While each of the entity types can be the target of a sentiment expression, the more interesting questions in this domain involve multiple entities, specifically, DRUG + DISEASE ("how effective is this drug for this disease?
Page 2, “Entity Relevance”
- (2010), working in the 'ignore relevance' mode, which (1) finds and labels all entities of the target type(s); (2) resolves all corefer-ences for the target entity type (s); (3) finds and labels all sentiment expressions, regardless of their relevance; and (4) provides dependency parses for all sentences in the corpus.
Page 3, “Relevance Algorithms”
- In the Financial corpus, COMPANIES are used as target entities and in the medical corpus, DISEASES, DRUGS and PERSONS are the entity types that are used as target entities.
Page 4, “Experiments”
See all papers in Proc. ACL 2014 that mention entity types.
See all papers in Proc. ACL that mention entity types.
Back to top.