Extracting Condition-Opinion Relations Toward Fine-grained Opinion Mining

Abstract

A fundamental issue in opinion mining is to search a corpus for opinion units, each of which typically comprises the evaluation by an author for a target object from an aspect, such as "This hotel is in a good location". However, few attempts have been made to address cases where the validity of an evaluation is restricted on a condition in the source text, such as "for traveling with small kids". In this paper, we propose a method to extract condition-opinion relations from online reviews, which enables fine-grained analysis for the utility of target objects depending the user attribute, purpose, and situation. Our method uses supervised machine learning to identify sequences of words or phrases that comprise conditions for opinions. We propose several features associated with lexical and syntactic information, and show their effectiveness experimentally.

Torii: An aspect‐based sentiment analysis system that can mine conditions

“…In the literature, Narayanan et al, Skeppstedt et al, Mausam et al, Mausam, Chikersal et al, and Nakayama and Fujii have worked on related proposals. Narayanan et al highlighted the problems of not dealing with conditions in the field of opinion mining.…”

“…We confirmed the previous claims with our experimental analysis. The only existing machine‐learning proposal was introduced by Nakayama and Fujii, who worked in the field of opinion mining in Japanese. They devised a model that is based on features that are computed by means of a syntactic parser and a semantic analyzer.…”

“…We used Chikersal et al's and Mausam et al's proposals as baselines. The proposal by Nakayama and Fujii was not taken into account because it is tightly bound to the Japanese language. Regarding our proposal, we evaluated the eight alternatives that result from combining the two alternatives to implement the encoder (GRU or BiGRU) and the two alternatives to implement the decoder (GRU or BiGRU) using both Softmax ( ϕ = soft ) and Sigmoid ( ϕ = sig ) as activation functions.…”

“…They generally attain good precision, but fall short regarding recall because there are a few usual ways to introduce conditions that can be easily modeled with user‐defined patterns, but too many unusual ways that cannot be easily modeled using such an approach (see the previous examples). There is only one machine‐learning proposal that attempts to deal with such unusual conditions. Unfortunately, it is bound with the Japanese language, it must be configured with several specific‐purpose dictionaries, taxonomies, and heuristics, and it was evaluated on a small data set with 3 155 sentences from hotel reviews.…”

A deep-learning approach to mining conditions

“…They generally attain good precision, but fall short regarding recall because the distribution of connectives that are used to introduce conditions is long-tail; that is, there are a few usual ways to introduce conditions that can be easily modelled using user-defined patterns, but too many unusual ways that cannot be easily modelled using such an approach. There is only one machine-learning proposal [8], which was intended to recognise as many patterns as possible in the long-tail distribution. Unfortunately, it is bound to the Japanese language, it must be customised with several specific-purpose dictionaries, taxonomies, and heuristics, mines conditions regarding opinions only, and it was evaluated on a small dataset with 3155 sentences that were sampled from hotel reviews.…”