Krishna Kumar T Chandrasekar (2016-09-20 13:45 - 14:30 in ZI-2042)
This talk examines the possibility for identifying successful fundraisers in online Health campaigns by analyzing the motivations provided by them. Choosing the right words improves the chances of raising money. A multi-gram language model was built based on the Naive Bayes classifier and Characteristic Rank words were found out for each of the different classes. This gives us an idea of the words that are to be used in the motivation and also the list of words that are to be avoided. This would be small step towards making a significant difference to the Health campaigns as the overall funds raised could be improved persistently owing to the enormous power of the social media.