Query Recommending Scheme : Implementations and Evaluation

Abstract

In general e-Discovery processing, most litigants depend heavily on lawyers for dealing with a series
of work required in the litigation process. Although there are other reasons that impel the lawyer to
do so, the most serious problem is that lawyer cannot know the detail information of litigant’s data
from the beginning. It causes misuse of keyword and then it makes poor results of evidences search,
so it will lower the efficiency of entire e-Discovery work. To solve these problems, we proposed the
concept of QRS through the ICACT 2014. In this paper, we introduce how we develop the QRS as
a system and evaluate the performance of QRS by experiments based on the two legal cases used in
TREC Legal Track.