Framework for language independent social media analysis platform to detect reactions on global topics
Takako Hashimoto, Supavadee Aramvith, Teeranoot Chauksuvanit, Yukari Shirota · 2014
Social media is recognized as the important collaborations in today's information oriented society. Especially, after the global topics like a severe disaster (e.g. the East Japan Great Earthquake, Thailand floods etc.), people all over the world post their thoughts/opinions to social media and their behaviors are sometimes influenced by the discussions on social media. Therefore, we would like to read and understand the contents on the worldwide SNS. Exploring topics on the social media globally is useful to gain a rich insight into the global social contexts. This paper proposes a framework for language independent social media analysis platform to detect the global topics. Our framework targets various countries' social media and extracts keywords from messages written by different languages. Then the framework translates keywords from a local language to English, so that we can understand meanings of keywords. Since our framework is based on keywords and it can extract topics from keywords using data mining techniques, once keywords are extracted, it can be language-independent. In this paper, our proposed framework for language independent social media analysis platform is described with examples of Thai social media analysis results.