Abstract
We introduce an extension to Multiple Classification Ripple Down Rules (MCRDR), called Contextual MCRDR (C-MCRDR). We apply C-MCRDR knowledge-base systems (KBS) to the Textual Question Answering (TQA) and Natural Language Interface to Databases (NLIDB) paradigms in restricted domains as a type of spoken dialog system (SDS) or conversational agent (CA). C-MCRDR implicitly maintains topical conversational context, and intra-dialog context is retained allowing explicit referencing in KB rule conditions and classifications. To facilitate NLIDB, post-inference C-MCRDR classifications can include generic query referencing – query specificity is achieved by the binding of pre-identified context. In contrast to other scripted, or syntactically complex systems, the KB of the live system can easily be maintained courtesy of the RDR knowledge engineering approach. For evaluation, we applied this system to a pedagogical domain that uses a production database for the generation of offline course-related documents. Our system complemented the domain by providing a spoken or textual question-answering alternative for undergraduates based on the same production database. The developed system incorporates a speech-enabled chatbot interface via Automatic Speech Recognition (ASR) and experimental results from a live, integrated feedback rating system showed significant user acceptance, indicating the approach is promising, feasible and further work is warranted. Evaluation of the prototype’s viability found the system responded appropriately for 80.3% of participant requests in the tested domain, and it responded inappropriately for 19.7% of requests due to incorrect dialog classifications (4.4%) or out of scope requests (15.3%). Although the semantic range of the evaluated domain was relatively shallow, we conjecture that the developed system is readily adoptable as a CA NLIDB tool in other more semantically-rich domains and it shows promise in single or multi-domain environments.
Highlights
Conversational Agents (CA) and Natural Language Interfaces to Databases (NLIDB) systems typically require the system developer/author to have high-level skills in constructing either complex semantic or syntactic grammars, or highly technical scripting languages to parse user utterances, as well as database querying languages such as SQL
One of the key features of C-Multiple Classification Ripple Down Rules (MCRDR) is the reduction in rule count due to post-inference querying and implicit reference to topical context
In this work our Contextual MCRDR (C-MCRDR) knowledge-base systems (KBS) modified standard MCRDR to facilitate the retention of topical and attribute-based context between inference requests, and to reduce the KB rule-count when used in domains with in situ databases by including generic querying bound by the intra-dialog context
Summary
Conversational Agents (CA) and Natural Language Interfaces to Databases (NLIDB) systems typically require the system developer/author to have high-level skills in constructing either complex semantic or syntactic grammars, or highly technical scripting languages to parse user utterances, as well as database querying languages such as SQL. This introduces a clear, unwarranted separation between the system author and a domain expert – ideally the domain expert should be able to author and maintain the knowledge required by the system, but it is unreasonable to expect domain experts to have high-level technical or linguistic analysis. In this paper we introduce C-MCRDR, which is a significant extension to MCRDR that facilitates constrained NL conversation via pattern-matching
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
