The NLP module divides input text into sentences and segments the word forms contained in each sentence into a string of morphemes. The segmented morphemes are grouped into syntactic units via syntactic analysis. Once syntactic units are constructed, expressions denoting named entities such as people, locations, and organizations are recognized. Then, association analysis is performed to identify tuples of <topic keyword, associated keyword>. Finally, sentiment polarities for topic keywords are determined through sentiment analysis. The results of the whole analysis are delivered in a time-series fashion using an application programmer’s interface (API) engine to accommodate various queries from users. The SOCIALmetricsTM system provides one of the most advanced solutions for the Korean language crawling and mining. Unlike English, Natural Language Processing in Korean is much more complicated. This is due to the fact that the Korean language exhibits characteristics of an agglutinative language and thus there has to be more than one morpheme in order to form a phrase. In the case of the English language, one morpheme is not separated as each word contains a single morpheme; however, the complexity of the Korean language is especially high as morphemes that construct a phrase have to be separated and each morpheme’s part of speech also has to be distinguished. In addition, a Korean word or phrase can carry a very different meaning when used in different linguistic contexts. In order to solve these challenges, SocialMetricsTM utilizes an extensive semantic classification dictionary that contains over 1 million words.