The project is a combination of various machine learning models that are applied to a personalised list of books a person has read and liked.

 

A dataset of classical books is analysed and then, based on the analysis of the person's preferences, a classical book can be recommended.

 

Some of the models used for the analysis of the datasets include LDA (Latent Dirichlet Allocation) which is a type of topic modelling, also SOM (self-organising maps), SVD (Singular Value Decomposition) and K-means.

 

The models are used for finding similarities within different texts/books, while the system makes use of content-based filtering. Through that, the modern and classical books that are the most similar in terms of topics can be determined.

 

The results of the analysis can be clearly seen through various graphs/visuals produced in the process.