Similar presentations:
Ai logic hirextra
1. HireXtra ASNA AI
HIREXTRAASNA AI
AI BASED RECRUITMENT
S O L U T I O N – J O B M AT C H O M E T E R
2. Applicant Skill analysing Algorithm using Artificial Intelligence ASNA.AI
APPLICANT SKILLA N A LY S I N G
ALGORITHM USING
ARTIFICIAL
INTELLIGENCE
ASNA.AI
3. Problem & Solution
Background : Approximately 800,000Recruiters face the same problem of
finding right candidates at faster pace
Problem - Recruiters spend hours, days,
weeks, sometimes months in finding
suitable Talent for Given Mandate
Solution -- Simple Automatic AI based
Candidate selection (from the LinkedIn
data, monster data)
AI Solution that works like human and
learns from previous Acceptance &
Rejections
PROBLEM &
SOLUTION
4. Big Data Technologies
MongoDB - It'sNoSQL opensource
cross-platform
document-oriented
database
Python – Fits into Big
data
Python :Eco-System,
high-level, general
purpose, interpreted,
dynamic programming
language suitable .
BIG DATA
TECHNOLOGIES
5. Data Scrubbing
Data CollectionVariable
Identification
Validation &
Cleaning
Addition of derived
variables
Preparation of
Model Set
Divide Training and
Testing dataset
DATA
SCRUBBING
6. Solution Framework
SOLUTION FRAMEWORKReceive JD –
Representation
of Data
DB- Job
Description
DB,
Results
displayed with
Fitment Score
Candidate DB
Results injected
as Feed Back:
Data PreProcessing
Learning's from
Results.
Hybrid Machine
Learning
Algorithm
The Domain
Knowledge
integrated
7. Basic Match Score
BASIC MATCH SCORE8. Career Progression Score
CAREERPROGRESSION
SCORE
9. AI/ML Implementation
A I / M LI M P L EMENTAT ION
10. AI Techniques - decision tree
AI TECHNIQUES - DECISION TREEDecision Tree - A decision tree is a natural and simple
way of inducing following kind of rules.
• If JD is for Data Scientist in Bangalore (Exp is 10) and
(Income is 12LPA) and (Nativity is KA) and
(Education is Ph.D.)(Previous Company is Start Up)
then he will be scored A+ ( example only )
• It is powerful and perhaps most robust .
Eg : The CV’s are scored
A+ 0.7 to 1
A – 0.5 to 0.7
B – 0.3 to 0.5
• Each CV has score based on their scores with Best
Match,
Career Progression Match, Social Media Grouping Score
• Learning from Acceptance and Rejection for
improvement
11. CV Matching Algorithm
CV MATCHING ALGORITHM12. Thank You
THANKYOU