Plone's related items functionality is a powerful feature but content managers mostly fail to do it (unless they are very committed and know their content very well). Here simserver comes to the rescue and assists in this task.

collective.simserver.core provides the common core functionality like an abstracted call interface, training of the corpus and indexing.

Plone communicates with the simserver via HTTP. For the Plone products to work, you will also need restsims https://github.com/cleder/restsims which is a small Pyramid wrapper around the Document Similarity Server itself. Simserver is built on Gensim. Gensim is a free Python framework designed to automatically extract semantic topics from documents, as efficiently (computer-wise) and painlessly (human-wise) as possible.

Conceptually, a service that lets you:

• Train a semantic model from a corpus of plain texts (no manual annotation and mark-up needed)
• Index arbitrary documents using this semantic model
• Query the index for similar documents (the query can be either an uid of a document already in the index, or an arbitrary text)

Digital libraries of (mostly) text documents. More generally, it helps you annotate, organize and navigate documents in a more abstract way, compared to plain keyword search.

• Enhance the UX by linking content to related content the user might also be interested in
• Easy way to tag documents
• SEO, improve the Page Rank of your site

See also: https://github.com/collective/collective.simserver.related More information at http://plone.org/products/collective.simserver/