Abstract:This paper investigates the question of the interpretability of fuzzy linguistic summaries, both at the sentence level and at the summary level, seen as a set of sentences. The individual sentence interpretability is examined as depending both on its representativity measured by a quality degree and on its linguistic expression. Different properties at the summary level are also discussed, namely their consistency, their non redundancy and the information they convey.
“…In this work, we apply the basic structure of linguistic summaries (2), evaluations expressed by the quantifier most of Kacprzyk and Zadrożny (2005), that is further parametrized in Hudec (2016) and aggregation functions (Beliakov et al, 2007), in order to explore the raised research questions. A review of the other types of linguistic summaries can be found in Lesot et al (2016), whereas a review of applicability can be found in Boran et al (2016). The solution of a summary is the validity or truth value of the evaluated quantified sentence, not a set of retrieved entities from (1).…”
Data users are generally interested in two types of aggregated information: summarization of the selected attribute(s) for all considered entities, and retrieval and evaluation of entities by the requirements posed on the relevant attributes. Less statistically literate users (e.g. domain experts) and the business intelligence strategic dashboards can benefit from the linguistic summarization, i.e. a summary like the most of customers are middle-aged can be understood immediately. Evaluation of the mandatory and optional requirements of the structure P 1 and most of the other posed predicates should be satisfied is beneficial for analytical business intelligence dashboards and search engines in general. This work formalizes the integration of aforementioned quantified summaries and quantified evaluation into the concept of database queries to empower their flexibility by, e.g. the nested quantified query conditions on hierarchical data structures. Next, this approach contributes to the mitigation of the empty answer problem in data retrieval tasks. Thus, the strategic and analytical dashboards as well as query engines might benefit from the proposed approach. Finally, the obtained results are illustrated on examples, the internal and external trustworthiness is elaborated, and the future research topics and applicability are discussed.
“…In this work, we apply the basic structure of linguistic summaries (2), evaluations expressed by the quantifier most of Kacprzyk and Zadrożny (2005), that is further parametrized in Hudec (2016) and aggregation functions (Beliakov et al, 2007), in order to explore the raised research questions. A review of the other types of linguistic summaries can be found in Lesot et al (2016), whereas a review of applicability can be found in Boran et al (2016). The solution of a summary is the validity or truth value of the evaluated quantified sentence, not a set of retrieved entities from (1).…”
Data users are generally interested in two types of aggregated information: summarization of the selected attribute(s) for all considered entities, and retrieval and evaluation of entities by the requirements posed on the relevant attributes. Less statistically literate users (e.g. domain experts) and the business intelligence strategic dashboards can benefit from the linguistic summarization, i.e. a summary like the most of customers are middle-aged can be understood immediately. Evaluation of the mandatory and optional requirements of the structure P 1 and most of the other posed predicates should be satisfied is beneficial for analytical business intelligence dashboards and search engines in general. This work formalizes the integration of aforementioned quantified summaries and quantified evaluation into the concept of database queries to empower their flexibility by, e.g. the nested quantified query conditions on hierarchical data structures. Next, this approach contributes to the mitigation of the empty answer problem in data retrieval tasks. Thus, the strategic and analytical dashboards as well as query engines might benefit from the proposed approach. Finally, the obtained results are illustrated on examples, the internal and external trustworthiness is elaborated, and the future research topics and applicability are discussed.
“…However, among other interesting fuzzy models, we would like to focus on linguistic summaries [7] [8], that combine the understandability of simplified natural language and the capacities of automatic learning and quality checking, the quality being understood in various senses. Their purpose is to sum up information contained in large volumes of data into simple sentences and the interpretability is at the core of the process [9]. The most generally used sentences, called protoforms, are of the form "Q B x s are A", where Q is a fuzzy quantifier representing a linguistic quantifier such as "most" or "a majority of", or, in the case of time series, a temporal indication such as "often" or "regularly", B is a fuzzy qualifier of elements x of the dataset to be summarised, sometimes omitted, and A is a fuzzy description of these elements called a summariser.…”
We show that many existing fuzzy methods for machine learning and data mining contribute to providing solutions to data science challenges, even though statistical approaches are often presented as major tools to cope with big data and modern user expectations of their exploitation. The multiple capacities of fuzzy and related knowledge representation methods make them inescapable to deal with various types of uncertainty inherent in all kinds of data.
“…One of their main advantage is human-consistency. In [18], Lesot et al investigate the interpretability of fuzzy linguistic summaries, both at the sentence level and at the group of summaries level.…”
Personalized linguistic summaries are developed with the use of protoforms in the sense of Yager and Kacprzyk. We discuss the construction and usefulness of such patientdependent and disease-state-dependent linguistic summaries that may be examplied as Most outgoing calls in mania state (disease period) are short compared to the calls recorded in the euthymia state (healthy period). Such linguistic summaries may become important features in the smartphonebased monitoring of the bipolar disorder patients and inform about the detected change in patient's state. The main advantage of the personalized linguistic summaries is their human-centricity. The performance of the proposed approach is illustrated with examples based on the real-life data collected within the observational study.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.