CV App

CV app is an AI tool for curriculum screening and people recruiting.

• Purpose: get skills of my candidates from CV screening (/metadata) or find best candidate for my job position (/candidate).
• Input & Output: it is implemented as a REST web-service (JSON).
• Info: this service uses Watson IBM Natural Language Understanding API.

Please read the API documentation below to get informations about each endpoint.

Installation

Install python 2.7 on a Linux distribution (tested on Ubuntu 16.04). Download package and unzip it. Run the following commands:

$ cd CV_app
$ sudo ./install_CVapp.sh requirements.txt
$ sudo python CV_server.py

To activate/deactivate the virtual enviroment, use:

$ source ~/dlenv/bin/activate
$ deactivate

POST: [ /candidate ]

• Purpose: Find best candidate for my job position.
• Input:
  • Folder path of my CV (directory) --> Required
  • Required skills for my job position (skills) --> Insert skills or label. Or both.
  • A label for this cluster of skills (label) --> Insert skills or label. Or both.
• Output:
  • Job offer label (position)
  • Candidate info (best_candidates)
    • Candidate Name (name)
    • Candidate Email (contact)
    • CV info (cv): - CV name (filename) - CV language (language) - CV skills (skills)

Request example (PYTHON):

import requests
import json

## HTTP DATA
my_headers={'Content-type' : 'application/json'}
my_url='http://127.0.0.1:80/candidate'

## INPUT DATA (JSON)
#Standard INPUT DATA (skills & label)
my_data={ "skills" : [ "science;computer science;artificial intelligence;;", "science;computer science;cryptography;;", "science;computer science;distributed systems;;", "science;computer science;information science;;", "technology and computing;operating systems;linux;;", "technology and computing;programming languages;java;;", "technology and computing;;;;", "technology and computing;technological innovation;;;", "technology and computing;software;databases;;", "technology and computing;operating systems;linux;;", "technology and computing;operating systems;unix;;", "technology and computing;programming languages;;;" ], "directory": "./pdf/", "label": "AI"  }

#Other type of INPUT DATA (no skills)
my_data={ "skills" : "", "directory": "./pdf/", "label": "AI"  }

#Other type of INPUT DATA (no label)
my_data={ "skills" : [ "science;computer science;artificial intelligence;;", "science;computer science;cryptography;;", "science;computer science;distributed systems;;", "science;computer science;information science;;", "technology and computing;operating systems;linux;;", "technology and computing;programming languages;java;;", "technology and computing;;;;", "technology and computing;technological innovation;;;", "technology and computing;software;databases;;", "technology and computing;operating systems;linux;;", "technology and computing;operating systems;unix;;", "technology and computing;programming languages;;;" ], "directory": "./pdf/", "label": ""  }

## HTTP REQUEST
r=requests.post(my_url,data=json.dumps(my_data), headers=my_headers)
r.json()

Request example (PYTHON):

import requests
import json

## HTTP DATA
my_headers={'Content-type' : 'application/json'}
my_url='http://127.0.0.1:80/metadata'

## INPUT DATA (JSON)
my_data={"directory": "./pdf/"}

## HTTP REQUEST
r=requests.post(my_url,data=json.dumps(my_data), headers=my_headers)
r.json()

License & Troubleshooting

This is a Manfrina web-app, released under GPL license. If it doesn't work is because server is being attacked by rebel robots. In that case, please fell free to contact me at: manfrina87ATgmail.com