Semantic search lets you search for documents not by keyword but by semantic content. This is particularly useful for fields in which a deeper understanding of the content of a document is required, for example legal or medical documents.
Find or compare
The problem with keyword based search
Most of nowadays search engines find documents by matching a set of keywords or phrases to their content - basically by performing a full-text search. For a lot of use cases this gives reasonable results. Nevertheless, there are many use cases in which this is not sufficient. This is especially true if a deeper understanding of the semantic content of the documents is required. One example are legal documents. Legal workers spend most of their time with searching (precedent) cases. Pure keyword based searches are not accurate enough for identifying the relevant cases, since they only incorporate the word statistic of a document. Thus, a method is needed that is able to understand its actual semantic content.
Search by meaning - not keyword
The answer to this problem is semantic search. Using the latest insights from NLP research, it is possible to train a Language Model on a large corpus of documents. Afterwards, the model is able represent documents based on their semantic content. In particular, this includes the possibility to search for documents with semantically similar content. Another possibility is to embed all documents in a 2-dimensional map, which makes it easy to visualize and explore the landscape of available documents.