1.Proposal Presentation

Semantic Matching of
candidates’ profile with job
data from Linkedln
• Introduction
• Research Questions
• Literature Survey
• System Architecture
• Project Scope
• Work division
• Future Goals
Table of Contents
• The Internet has already evolved into
the primary medium for recruitment
and employment processes. Most of
recently filled positions in US are the
result of online job posting and
according to most forecasts this share
is likely to increase further.
• General websites(e.g FreeResumeWeb) say they collect the hot jobs
from different job sites provide more job information. The
International Association of Employment Web Sites reports that there
are more than 40,000 employment sites serving job seekers,
employers and recruiters worldwide.
• Many professionals use social network LinkedIn as a job-hunting
tool, and many companies use it for hiring.
• Friends
• Co-workers
• But, the amount of job data
returned in the results is huge,
which means that the job-hunters
need more time to scan through
those jobs.
• The “Big Data” is too powerful
for individual to use?
• LinkedIn
• Facebook,Twitter
• Career Board
Web Sites
Research Questions
• Company
• Career websites
• Example: “software developer”
searched on LinkedIn
• Job searching
• Job matching
Current Situation
• Latest posted Job vacancies will be found first, what about the
“older” ones?
• So much of the job information after searching in different job
portals, how to filter those?
• Just adding more search constraints does not help.
• How to deal with the overlapping information amongst the
different datasets (LinkedIn, Glassdoor etc)?
• If you have no professional connections in LinkedIn, the
search results will be different, which one is good for the job
"People are impatient, even lazy, " Dickey-Chasins observes. "They don't want to spend 10,
20 or even 30 minutes filling out forms for an uncertain
Different matchmaking approaches exist in the literature which can be used for matching
individuals to job requirements.
For example, typical text-based information retrieval techniques such as database querying and
similarity between weighted vectors of terms have been used in previous works (Veit et al, 2006).
(Liu and Dew, 2004) presents a system which integrates the accuracy of concept search with the
flexibility of keyword search to match expertise within academia.
(Colucci et al, 2003) proposes a semantic based approach to the problem of skills finding in an
ontology supported framework. They use description logic nferences to handle the background
knowledge and deal with incomplete knowledge while finding the best individual for a given task
or project, based on profile descriptions sharing a common ontology.
Approaches for calculating the structural similarity between instances on the basis of ontologies
have also been considered. (Bizer et al, 2005) and (Mochol et al, 2007), for example, present a
scenario for supporting the recruitment process with semantic web technologies which uses (Zhong
et al, 2002)’s similarity measure to evaluate the degree of match between job offers and applicants.
Semantic Matching for Job Requirement: An ontology based Hybrid approach
The Impact of Semantic Web Technologies on Job Recruitment Processes
Literature Survey
Link different
Job description
Matching the
job vacancies
with the jobseeker’s profile
Our Idea
of the
System details - Data Set
User Job profile
Job description
Sequence Diagram
Problem in hand: the calculation of the degree of
semantic similarity between an applicant's profile and job requirements.
We use the term competence to refer to a skill along with a level of
We do not include skills related to specific tools and technologies.
We define a competency level in terms of the required level of knowledge
and years of experience.
We distinguish between four levels of knowledge: basic, intermediate,
advanced, and expert knowledge.
The years of experience is specified as the minimum number of years
The similarity between two concepts is determined by the distance between
them, which reflects their respective positions in the concept hierarchy.
Based on set of rules:
Give a high score for sibling instances, a smaller score for
cousin instances, with plus or minus points if the required competency
level is satisfied
(extra points for experience at or above required competency level etc)
Semantic Matching
• Building an ontology representing user’s job profile.
• Getting the job postings data from LinkedIn and other websites.
• Using Yago2 to integrate the job posting data with the location information
for visualization.
• Using Google Maps API for map visualization.
• Using Jena as the Java Framework to build the semantic web application.
• Implementing the Semantic Matching Engine.
Project Scope
• Ontologies Design
- Sarabpreet, Ting
• Dataset building
- Ting
• Yago2 integration.
- Sarabpreet
• Designing and implementing the Web Interface
- Sarabpreet, Ting
• Matching Engine
- Sarabpreet, Ting
Work Division
• Integrating various datasets – Linkedin, Glassdoor,
DawgLink, Monster etc and removing the information
• Adding the Community feature.
• Extend the functionality to provide similar matching
capabilities to employers looking for suitable candidates.
Future Goals
Thank You!