What is xml based keyword search?

What are the differences between semantic based search and keyword based search?

• Answer:

  Fundamentally, it's the difference between "give me what I said" and "give me what I want." (h/t ) Semantic search seeks to improve search accuracy by understanding searcher intent and the contextual meaning of terms as they appear in the searchable dataspace, whether on the Web or within a closed system, to generate more relevant results. Keyword search is a type of search that looks for matching documents that contain one or more words specified by the user. For example, I am looking for (association) football scores for the English Premier League. I am very vague though, the only search query I can come up with is "football."   Bing, which is still pretty keyword-based, gives me this: Lots of links about football. Bing has found popular sites that have football as a keyword, and also does things like cross-references this based on my location, and a few other factors clear to it. It's not making too many guesses. I'll probably find something there I like, or I will refine my search, probably by adding "scores" or "epl" or something to the query. Google, on the other hand, gives me this: This is a totally different. There are links, but there is also a scoreboard and a map of football related locations around me (London). It's inferred from my location that I must be interested in English football, and because it has access to a database of scores, it can pull and display these even though I've not requested them. Essentially, it's guessed what I'm looking for, using a bunch of factors. It has associations between football and things like "stadiums (and their locations)", "clubs", "scores", "news" etc. to connect my search query with and so present data that could be relevant to me. These associations are popular ones, and likely manually created. Most people searching for football are looking for scores, so the association is relevant to most searchers (but not all!). True semantic search would see the a search engine able to create these associations on its own, essentially machine-learning. Right now, the associations are made by an engineer, and the system cannot really organically learn and implement them.