In Search of Accurate Approach for Opinion Mining

Singh, V. K., Mukherjee, M., Mehta, et al. “Opinion Mining from Weblogs and Its Relevance for Socio-political Research“, 2012.

1. using a text classifier (such as Naive Bayes, SVM or kNN) that takes a machine learning approach to categorise the documents;
2. computing semantic orientation of documents based on aggregated semantic orientation values of selected opinionated POS tags extracted from the document.

Bollen, J., Pepe, A., & Mao, H. “Modeling public mood and emotion: Twitter sentiment and socio-economic phenomena, 2010.

1. Definition of data and mood assessment instrument.
2. Data cleaning, parsing and normalisation.
3. Time series production: aggregation of POMS mood scores over time.
4. Comparison of produced mood time series to socio-economic indicators.