Harnessing Twitter ‘Big Data’ for Automatic Emotion Identification

Describes some of the methods used in Cognovi’s emotion identification algorithms. Please note that Cognovi Labs’ current emotion extraction algorithm is based on an enhanced approach, which generates significantly higher accuracy levels than described in this paper. The algorithm is kept confidential and has never been published. User generated content on Twitter (produced at an enormous rate of 340 million tweets per day) provides a rich source for gleaning people's emotions, which is necessary for deeper understanding of people's behaviors and actions. Extant studies on emotion identification lack comprehensive coverage of "emotional situations" because they use relatively small training datasets. To overcome this bottleneck, we have automatically created a large emotion-labeled dataset (of about 2.5 million tweets) by harnessing emotion-related hashtags available in the tweets. We have applied two different machine learning algorithms for emotion identification, to study the effectiveness of various feature combinations as well as the effect of the size of the training data on the emotion identification task. Our experiments demonstrate that a combination of unigrams, big rams, sentiment/emotion-bearing words, and parts-of-speech information is most effective for gleaning emotions. The highest accuracy (65.57%) is achieved with a training data containing about 2 million tweets. Direct link to publication: Harnessing Twitter ‘Big Data’ for Automatic Emotion Identification