Question Answering over Pattern-Based User Models

Abstract

In this paper we present an ontology-driven framework for natural language question analysis and answering over user models (e.g. preferences, habits and health problems of individuals) that are formally captured using ontology design patterns. Pattern-based modelling is extremely useful for capturing n-ary relations in a well-defined and axiomatised manner, but it introduces additional challenges in building NL interfaces for accessing the underlying content. This is mainly due to the encapsulation of domain semantics inside conceptual layers of abstraction (e.g. using reification or container classes) that demand flexible, context-aware approaches for query analysis and interpretation. We describe the coupling of a frame-based formalisation of natural language user utterances with a context-aware query interpretation towards question answering over pattern-based RDF knowledge bases. The proposed framework is part of a human-like socially communicative agent that acts as an intermediate between elderly migrants and care personnel, assisting the latter to solicit personal information about care recipients (e.g. medical history, care needs, preferences, routines, habits, etc.).

Question Answer System: A State-of-Art Representation of Quantitative and Qualitative Analysis

“…Many KBQA systems like Nicula et al [99], Le et al [100], Shin et al [101] use standard dependency tree generation methods. Phase structure grammar [102] and feature-based grammar [103] have been used to generate the dependency tree, but the most prominent method uses the parsing tools like TALN [104] and Stanford Parser [105][106][107]. Hu et al [95] proposed the Valuable Dependency Parser, which uses the Stanford Parser for initial parsing, and then a few tags are prioritized for query generation.…”

“…To map the extracted entities to entities in KB, many QA systems have used the existing tools like Dbpedia spotlight [106,109,110,[117][118][119][120], S-MART [114]. The similarity measures like Jaro-Vinkler [101], UMBC+LSA [104], and Siamese LSTM [121] have also been used for mapping the entities. Few QA systems have also used approaches based on NN [122], Hierarchical RNN [16], BiLSTM [123], and BERT [124,125].…”

Converness: Ontology‐driven conversational awareness and context understanding in multimodal dialogue systems

“…(Pragst et al, ). Question answering for retrieving responses relevant to the topic of discussion and the needs of dialogue manager. As we describe in Section , the integrated platform supports the retrieval of information that is either relevant to behavioural aspects and profile information of users (Meditskos, Dasiopoulou, Vrochidis, Wanner, & Kompatsiaris, ) or to generic information retrieved from web resources (Moumtzidou et al, ). Language generation for communicating information to the users. The verbal communication capitalises on the ontological representations returned from question interpretation, following the inverse cascade of processing stages described in language analysis (Mille, Carlini, Burga, & Wanner, ). Avatar for the agent's nonverbal appearance.…”

“…• Question answering for retrieving responses relevant to the topic of discussion and the needs of dialogue manager. As we describe in Section 6.1, the integrated platform supports the retrieval of information that is either relevant to behavioural aspects and profile information of users (Meditskos, Dasiopoulou, Vrochidis, Wanner, & Kompatsiaris, 2016) or to generic information retrieved from web resources (Moumtzidou et al, 2016).…”