Build me traceability Information Retrieval Methods

I'm a researcher attached with an academic institution. My research focus: Trace semantic similarity between documents. For example, the semantic similarity score of -- content of a test case traceable to --- a requirement. I am interested in using the following Information Retrieval Methods: (1) Latent Dirichlet Allocation (LDA) (2) Jensen-Shannon (JS) (3) Vector Space Model (VSM) (4) Latent Semantic Indexing (LSI) (5) Normalized Google Distance (NGD) (6) Pointwise Mutual Information (PMI)