Geek asks and answers

How to create the best outsourcing using machine learning

I am creating an autosuggest function.

I have a data set of about 23 thousand records, in this data words and combinations of several words, i.e. London College of the Arts and London.

I want to write an autocomplete function that takes into account letters that have already been entered, for example, "LO", as well as some other data points that I have:

  • a place,

  • day of the year

  • number of previous requests

  • collected subjects - say, the London College of Art has 20,000 students, which makes it higher than the London School of Economics, where there are only 5,000, etc.

I suppose that from the preliminary studies I need to start studying the machine learning technique, but it’s still hard for me to see what SVN, NN etc. is, what is best applied to this problem. It is difficult for me to find any practical examples of using these algorithms for such problems, any help or pointers to good resources is very appreciated.

+5
source share
1 answer

Take a look at Apache Solr (this may be redundant for your needs for an automatic offer only), but it offers a pretty decent OOTB autosuggest, without having to create from scratch. Since it offers a RESTful interface, you can integrate it into any language of your choice.

, , .. Solr - , , , .

+2

More articles:


All Articles