Cosmopolitan Mumbai, Orthodox Delhi, Techcity Bangalore: Understanding City Specific Societal Sentiment

The paper reports work on investigating societal sentiment using the Schwartz values and ethics model, and applying it to social media text of users from 20 most populous cities of India to represent geo-specific societal sentiment map of India. For the automatic detection of societal sentiment we propose psycholinguistic analysis, that reveals how a user’s social media behaviour and language is related to his/her ethical practices. India is a multi-cultural country, values and ethics of each Indian are highly diverse and dependent on the region or society s/he belongs to. Several experiments were carried out incorporating Linguistic Inquiry Word Count analysis, n-grams, topic modeling, psycholinguistic lexica, speechacts, and non-linguistic features, while experimenting with a range of machine learning algorithms including Support Vector Machines, Logistic Regression, and Random Forests to identify the best linguistic and non-linguistic features for automatic classification of values and ethics.