Machine-Learning-for-textual-data

In this project we analyse the bible text and other religious texts using machine learning for textual data. From this analysis we expect to see different relevant topics and contexts between the books of the bible and between the different religious texts. This analysis is structured as follows: In a first step, a thorough analysis of the bible is done to understand different topics within the books and to see a development of the topics over time. We expect to see that certain topics where clearly more relevant in the old testament than in the new testament and vice versa. For the topic analysis in this paper, we will use the Latent Dirichlet Allocation (LDA) and the Word2Vec algorithm is used to understand the context of certain words. This allows to capture differences in the perception of one specific word in the old vs. new testament. We also apply a sentiment analysis to identify the sentiment evolvement over time. In a second part we will then include other religious texts such as the Quran and others. We will perform an anlysis over five texts to find similarities and patterns. We use the same methods as in PART I, namely LDA, Word2vec and sentiment.