Online consumer communities serve as a useful means to gather consumer opinions on various relevant topics. These communities are particularly beneficial for conducting consumer studies in multiple countries, as most platforms come equipped with machine translation functions. However, it is important to note that some consumer insights may not be accurately captured by mechanical translations. To overcome this limitation, we interpret the implied meanings behind consumers’ statements to gain deeper insights.
The Context
A global research agency wanted to conduct an online user community ( we also call it Online Bulletin Boards) research among gamers to explore the potential of new games in Japan. They want to know what games they currently play and what they think of concepts of new game prototypes.
The Challenge
Understanding jargon, implicated phases, and incomplete sentences
While machine translations translate what users chat about on the bulletin board in real time, the machine often mistranslates what the users meant to say. For example, “Why were they talking about locust (“イナゴ”: Inago in Japanese) here?” “What is “Nuruge?” The machine translation translated “Nuruge” (ヌルゲー in Japanese) as Nuruge, which does not make any sense. Furthermore, the machine translations fail to translate fragmentary phrases of incomplete sentences.
The Solution Approaches
#1. Interpret the meanings with context
In user communities, especially among gamers, sometimes users use jargon in their communications to share their feelings and interests. Therefore, we interpreted the contents on the community board by examining wordings and phrases used often in the gaming communities. For example, “Inago” is jargon among gamers, not the meaning of locust, but meaning “fans who do not have a specific genre, jump into seasonal games that are popular at that time, make a noise, and then move on to the next trendy genre.” Likewise, “Nurge” means “Nama-nurui game (Game that is too easy to play).”
#2. Moderate discussions by reading between the lines
We also moderated the online discussions by reading their emotions and opinions on what they wrote. For example, a user misunderstood task completions relevant to their gaming skills and hesitated to clarify them with the moderator. In this case, the moderator quickly jumped into the forum and explained the task requirements.
The Outcome
We presented the results of the research both verbally and in written format in English. The outcomes of the study were positively received and utilized in the creation of new games.
