Sentiment Analysis algorithms work by referring external resources where the positive and negative polarity of each word is considered. These external words, for which the polarity is predetermined, are known as lexicons. There are several lists of lexicons available and each one focuses on the polarity of a given word in a particular context. For example, consider the following couple of sentences:
The play has an unpredictable plot
This car's steering is unpredictable
The first usage of unpredictable indicates that the play is good. If this statement is found on an online portal that lets users rate plays, then this phrase can be taken as a positive feedback for the play involved. On the other hand, the usage of unpredictable in the second sentence means that driving the car could be dangerous because the steering wheel is unpredictable in its functioning. Thus, if we use a general-purpose lexicon that has a high negative polarity for...
