Cruncher Implementation

Web service interface documentation can be found here

Job Description examples can be found here

The approach following is a very simple word counting approach where we count the number of times each word is present in the document and use that to define wish jobs are related with.

To implement this the following libraries were used:

• Spring-boot
• This library is used to implement all the MVC, security, serialization, connection with database
  
• Eh-Cache
• This library is used to cache information for security, and in the future also some relevant information on resumes
• Guava
• Mainly for tests information
• JUnit
• For the testing framework
• Mockito
  
• For mocking in tests
• Tika
• To parse the resume files
  

Video that describes the current implementation of the cruncher and how we are doing the connection with the application itself:



Eventually the future will bring new version of this using algorithms like Naive Bayesian Classification or even Mahout.