Implementing Inductive Concept Learning For Cooperative Query Answering

A Replication Scheme for Multiple Fragmentations with Overlapping Fragments

“…Query generalization has long been studied in flexible query answering and machine learning (see the seminal article [44]). Query generalization at runtime has been implemented in the CoopQA system [45,46] by applying three generalization operators to a conjunctive query; while two of them (Dropping Condition and Goal Replacement) are purely syntactic operators, the third called Anti-Instantiation (AI) introduces a new variable and might be semantically restricted to avoid overgeneralization; this is what we do in this paper by obtaining fragmentations based on a clustering of the active domain of a relaxation attribute. More precisely, AI replaces a constant (or a variable occurring at least twice) in a query with a new variable y.…”

“…Moreover, query generalization at runtime (as for example implemented in [46]) is highly inefficient. That is why our clustering-based fragmentation preprocesses data into fragments of closely related values (with respect to a relaxation attribute).…”

Using a Dialogue Manager to Improve Semantic Web Search

“…Farah Benamara presents several works in this area: in (Benamara, 2004b), presents a logic-based model for accurate generation of intentional answers using a Cooperative QA system; in (Benamara, 2004a), presents a proposal for construction of a Logic-Based QA system, WEBCOOP, that integrates knowledge representation and advanced strategies of reasoning to generate cooperative answers to web queries. More recently, in (Bakhtyar, Dang, Inoue, & Wiese, 2014), the authors present an implementation of conceptual inductive learning operators in a prototype system for cooperative query answering, which can also be used as a usual concept learning mechanism for concepts described in first-order predicate logic.…”