Quality Assessment in Linguistic Summaries of Data

Castillo-Ortega, Rita; Marín, Nicolás; Sánchez, Daniel; Tettamanzi, Andrea G. B.

doi:10.1007/978-3-642-31715-6_31

Cited by 17 publications

(8 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The common problem is combining (aggregating) these measures. A relation ≤ Q [3] having less quality than generally works when LS i is better than LS j by all measures. Otherwise it is not an easy task, because measures might be conflicting or partially redundant.…”

Section: E Discussionmentioning

confidence: 98%

“…Castillo-Ortega et al [3] examined quality by four measures: coverage and brevity (shortness) (which are equivalent to two discussed in [4], [16]), specificity and accuracy. The common problem is combining (aggregating) these measures.…”

Section: E Discussionmentioning

confidence: 99%

“…An indispensable part of data summarization related to territorial units' data should be quality measures (these measures are discussed in e.g. [3,4], [16]) in order to exclude irrelevant summaries (low validity, low data coverage, relations between outliers and the like).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Linguistically summarizing hierarchical data: The case of statistical territorial data

Hudec

2015

2015 16th IEEE International Symposium on Computational Intelligence and Informatics (CINTI)

View full text Add to dashboard Cite

Linguistic summaries are suitable for summing up information from the large amount of data by simulating human reasoning. Summarized information helps users to focus on a relevant higher hierarchical level by asking short questions about data on respective lower levels. Territorial units are organized in a hierarchy from the top level to the lowest level. Two main parts of linguistic summaries are predicate (consisting of summarizer and restriction) and relative quantifier, both expressed as fuzzy sets. Therefore, summarized information is expressed as a quantified sentence of natural language. Quality measures ensure that only summaries which bear relevant information and are not calculated on outliers are valuable. Furthermore, benefits of linguistic summaries are demonstrated on illustrative examples on real territorial units' data. Finally, perspectives for further research are touched.

show abstract

Section: E Discussionmentioning

confidence: 98%

Section: E Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Linguistically summarizing hierarchical data: The case of statistical territorial data

Hudec

2015

2015 16th IEEE International Symposium on Computational Intelligence and Informatics (CINTI)

View full text Add to dashboard Cite

show abstract

“…For example, in [95], Díaz-Hermida et al explore several theoretical aspects such as the use of semi-fuzzy quantifiers to model quantified sentences and the description of some generic methods for pattern detection. Furthermore, [96][97][98][99][100][101] explore several and mostly convergent evaluation criteria, such as the data coverage percentage, the sentence truth degree, and other inspired by the conversational maxims in the field of human communication [102], including the relevance, specificity, ambiguity or length of the description. In fact, when referring to criteria, it can be stated that there is a solid consensus about which characteristics of a linguistic description can be useful in the task of evaluating and ranking candidate descriptions in an objective way.…”

Section: Theoretical Workmentioning

confidence: 99%

On the role of linguistic descriptions of data in the building of natural language generation systems

Ramos-Soto

Bugarín

Barro

2016

Fuzzy Sets and Systems

View full text Add to dashboard Cite

“…This means that there is in general no such thing as the optimum or best referring expression for a given object, and the referring expression generation is a multiobjective optimization problem, as explained in [30,31] for complete linguistic descriptions.…”

Section: A Multiobjective Problemmentioning

confidence: 99%

The Role of Graduality for Referring Expression Generation in Visual Scenes

Gatt

Marín

Portet

et al. 2016

Communications in Computer and Information Science

Self Cite

View full text Add to dashboard Cite

Abstract. Referring Expression Generation (reg) algorithms, a core component of systems that generate text from non-linguistic data, seek to identify domain objects using natural language descriptions. While reg has often been applied to visual domains, very few approaches deal with the problem of fuzziness and gradation. This paper discusses these problems and how they can be accommodated to achieve a more realistic view of the task of referring to objects in visual scenes.

show abstract

Quality Assessment in Linguistic Summaries of Data

Cited by 17 publications

References 7 publications

Linguistically summarizing hierarchical data: The case of statistical territorial data

Linguistically summarizing hierarchical data: The case of statistical territorial data

On the role of linguistic descriptions of data in the building of natural language generation systems

The Role of Graduality for Referring Expression Generation in Visual Scenes

Contact Info

Product

Resources

About