Quran Sentiment Analysis using Natural Language Processing

Sentiment Analysis is the automated process of analyzing text data and sorting it into sentiments. In this article, we use the IBM Watson Tone Analyzer to assess the dominant emotions and language style in every chapter of the Quran. Emotion categories are joy, sadness, anger, and fear, whereas language style categories are analyitcal, confident, and tentative. For instance, the Tone Analyzer can assess the tone of Donald Trump’s tweets.

The example below is a screenshot of the Tone Analyzer interface applied to our case study: the Quran. Key sentences are automatically classified by the program. A score number between 0 and 1 is assigned to each classification result. A score number above 0.75 suggests that the given emotion or language style is very likely to be present in the text. A score number below 0.5 suggests that the given emotion or language style is unlikely to be present in the text.

In the following sections, we present the results of the analysis for all chapters of the Quran.

Results

The x-axis represents the Quran chapters chronologically ranked and the y-axis represents the estimated score of dominant emotions. We observe that joy (in orange) is the dominant emotion throughout the Quran.

The x-axis represents the Quran chapters chronologically ranked and the y-axis represents the estimated score of dominant language styles. We observe that the analytical style ( in orange) is the dominant language style throughout the Quran.

For clarity, the previous plots are broken down into several individual plots representing the occurrence of each emotion and language style.

Emotions Breakdown

Language Style Breakdown