Question-Answering for Agricultural Open Data - Transactions on Large-Scale Data-and Knowledge-Centered Systems - page 15

Question-Answering for Agricultural Open Data

B

Takahiro Kawamura 1 , 2(

) and Akihiko Ohsuga 2

1 Corporate Research & Development Center,
Toshiba Corporation, Tokyo, Japan
[e-mail]
2 Graduate School of Information Systems,
University of Electro-Communications, Tokyo, Japan

Abstract. In the agricultural sector, the improvement of productivity
and quality with respect to such attributes as safety, security and taste
has been required in recent years. We aim to contribute to such improve-
ment through the application of Information and Communication Tech-
nology (ICT). In this paper, we first propose a model of agricultural
knowledge by Linked Open Data (LOD) with a view to establishing an
open standard for agricultural data, allowing flexible schemas based on
ontology alignment. We also present a semi-automatic mechanism that
we developed to extract agricultural knowledge from the Web, which
involves a bootstrapping method and dependency parsing, and confirmed
a certain degree of accuracy. Moreover, we present a voice-controlled
question-answering system that we developed for the LOD using tripli-
fication of query sentences and graph pattern matching of the triples.
Finally, we confirm through a use case that users can obtain the nec-
essary knowledge for several problems encountered in the agricultural
workplace.

· ·

Keywords: Question-answering system Linked Open Data Agricul-

ture

1 Introduction

Concern about food shortages affecting people in various parts of the world has
been rising in recent years. However, since the expansion of the cultivated area
is subject to constraints, it is necessary to improve productivity. On the other
hand, the improvement of quality (safety, security, taste) is required in order to
raise the incomes of farming households. In these circumstances, we aim to con-
tribute to agricultural production by applying ICT techniques. Unlike produc-
ers involved in industrial production, most farmers, except for exemplary ones,
are often confronted by unanticipated questions pertaining to several activities
ranging from planting to harvesting, since agricultural work depends on compli-
cated environmental factors. Thus, research[ 1 ] has been conducted with a view
to developing a search engine to find previous problems similar to current prob-
lems. However, searching the Internet using a smartphone or a tablet PC on site
c