Guided Conversational Agents and Knowledge Trees for Natural Language Interfaces to Relational Databases

estonianmelonΤεχνίτη Νοημοσύνη και Ρομποτική

24 Οκτ 2013 (πριν από 3 χρόνια και 7 μήνες)

81 εμφανίσεις

Guided Conversational Agents and Knowledge Trees for
Natural Language Interfaces
to Relational Databases

Majdi Owda
, Zuhair Bandar, and Keeley Crockett.

The Intelligent Systems Group, Department of Computing and Mathematics,

The Manchester Metropolitan Un
iversity, Chester Street,

Manchester, M1 5GD, UK.


Developing reliable natural language interfaces to relational databases (NLI
RDBs) is a
problematic task, since it is related to the ultimate purpose of artificial intelligence, natural
anguage understanding.
A considerable amount of literature has been published on the area
of NLIDBs. Androutsopoulos
(Androutsopoulos, Ritchie et al. 1995)

defined four main
approaches to NLIDBs: Pattern

an intermediate language


, and Semantic
Grammar Based

Currently there is some
ongoing work on

NLIDBs which does not fall into the categories defined by Androutsop
such as Precise
(Popescu, Etzioni
et al. 2003)
, Step
(Minock 2005)

A conversational agent is human
computer dialogue system that interacts with the user turn
y turn using natural language.
Yager (Yager 2006)

defines the idea of a knowledge tree as a
mechanism providing a global framework for the mobilization of a knowledge base in
response to a specifi
c query. Our definition of a knowledge tree is a tree where the knowledge
is organized in a hierarchical structure based on the expert knowledge which has been
extracted and developed by a knowledge engineer. Hence the knowledge trees serves as a
road map
for the conversation flow in a specific domain.


This abstract proposes a new approach for creating conversation
based natural language
interfaces to relational databases by combining goal oriented conversational agents and
knowledge trees. Go
al oriented conversational agents have proven their capability to
disambiguate the user’s needs and to converse within a context (i.e. specific domain) by
lowing for dialogue interaction,

Whereas k
nowledge trees used to structure the domain
knowledge to
cover common queries in a particular domain.
Knowledge trees also work as a
road map for the conversational agent dialogue flow towards the goal (i.e. query generation
from natural language).

Results and

The proposed approach reduces the com
plexity of building systems by using the knowledge
trees to structure and maintain domain knowledge, which is more effective to use than using
problematic syntax and semantic grammars. The developed prototype system shows excellent
performance on common qu
eries (i.e. queries extracted from expert by a knowledge
engineer). The user will have a friendly interface that can converse with the relational


Androutsopoulos, I., G. D. Ritchie, et al. (1995). "Natural language

interfaces to databases
: 29

Journal of Langauge Engineering

Minock, M. (2005). A Phrasal Approach to Natural Language Interfaces over Databases.
, Springer Berlin / Heidelberg.
Natural Language Processing and Information Systems
ume 3513/2005:

Towards a theory of natural language interfaces to
Popescu, A., O. Etzioni, et al. (2003).
. 8th international conference on Intelligent user interfaces, Miami, Florida,
USA, ACM Press, New York, NY, USA.

Yager, R. (2006).
"Knowledge trees and protoforms in question
answering systems: Special
Topic Section on Soft Approaches to Information Retrieval and Information Access
: 550

J. Am. Soc. Inf. Sci. Technol.
on the Web."