Sentiment Analysis (also known as opinion mining) refers to the use of natural language processing, text analysis and computational linguistics to identify and extract subjective information in source materials. – Wikipedia.org
There are two models widely used in automatic sentiment analysis.
The model most frequently used in commercial applications is based on linguistic resources and the other is based on machine learning. The big issue with sentiment analysis based on linguistic resources is that it’s difficult to (consistently) predict the many ways sentiment can be expressed.
Machine learning relies on the ability of a machine (computer) to learn the language used for expressing sentiment regardless of how “good” or “normal” the language is. The challenge here revolves around knowledge domains.
If a machine used a training corpus based on digital camera reviews, it’ll be pretty inaccurate when applied to reviews of fountain pens. For machine learning to work well, there needs to be some serious investment (by humans) in creating robust, contextually rich training corpora across all domains.
Understanding the context, both social and cultural, of the language and phrases being analyzed is incredibly important to accurately tailor sentiment predictions. This is especially true when dealing with sarcasm and other types of derisive or ironic language.
So, no matter what type of sentiment analysis your application uses, keep in mind that the results won’t be perfect.
- “Opinion Mining and Sentiment Analysis” – Bo Pang and Lillian Lee
- “Stanford researchers to open-source model they say has nailed sentiment analysis” – GigaOm
- “Creating a Sentiment Analysis Model” – Google Developers
- “Strange Uses for Sentiment Analysis” – Smart Data Collective
- “On Social Sentiment and Sentiment Analysis” – brnrd.me
- “Why is Sentiment Analysis Hard?” – idibon
- “50 Top Tools for Social Media Monitoring, Analytics, and Management” – Social Media Today
- “Trade the Sentiment – Using Sentiment and Technical Analysis to Trade the Market” – Eric D. Brown