Image Based Tool for Level 1 and Level 2 Autistic People
Job Seeker
1. Mange Chen (machen@indiana.edu)
Department of Computer Science, Indiana University, Bloomington, IN USA
Job Seeker
Introduction
The planner of Job Seeker is to build a
business and employment-oriented database
that operates via external party developer API
such as Wikipedia.
Design Strategy Conclusion & Future Work
References
Acknowledge
[1]. Lin, Yiou, Hang Lei, Prince Clement
Addo, and Xiaoyu Li. Machine Learned
Resume-Job Matching Solution. Computation
and Language. ArXiv.org, 26 July 2016.
Web.
[2]. Hirst, Tony. "OUseful.Info, the Blog..."
OUsefulInfo the Blog.
Https://blog.ouseful.info/, 30 Apr. 2016.
Web. 05 Dec. 2016.
Results & Analysis
Implementation
Objective
The basic functionality of Job Seeker allows
(workers and employers) to find the suitable
job or applicants by searching key words of
skills.
•Users can observe the degree of correlation
through a chart of skills.
•Users can find jobs, people and business
opportunities by searching key words of their
skills such as Ruby.
•Employers can find the most qualified
applicants.
Job
Seeker
Step One: Use Python to crawl skills
data (pages, subcategories, categories).
Step Two: Use NetworkX package in
Anaconda to generate graphs.
• These solutions are typically driven by
manual search-based rules and pre-defined
keyword weights, which results in an
inefficient and frustrating search
experience.
• The job of searching through online
matching engines is now very prominent
and is beneficial to job seekers and
employers to extract information directly
from resumes and vacancies.
• By search single key word of skills, we
can find out other skills related to it. This
information provides chance for workers
to increase their competitiveness.
This work is supported by the Undergraduate
Research of Computing (UROC) in Indiana
University Bloomington.
Thank Mohsen Sayyadiharikandeh for
comments that greatly improved the
manuscript, and Cassidy Wichowsky for
sharing her pearls of wisdom with me during
the research.
Figure 2. This is a example of the sort of thing we can
get out for a search seeded on skills associated with
Computer Science such as Ruby.
The to-do list is automated by adding some network
statistics to the NetworkX step and possibly a first
pass layout.
Figure 1. The simple connection between pages,
subcategories and categories. 1 is the main
category contains the subcategories 2 and 0.
Both 2 and 0 are subcategories. 3 is the page
which belongs to subcategory named 2.
• In this poster, I have considered the
matching problem of skills and
proposed a solution by using ensemble
methods.
• In the future, through the website to
update more information, job seekers
solutions can be extended by including
location information, professional
skills and requirements to expand.
Figure 4. By way of demonstrating how the recipe described in
Visualizing Related Entries in Wikipedia Using Gephi can easily be
turned to other things, here’s a map of how different computer
programming languages influence each other according to
DBpedia/Wikipedia. [2]
Figure 3. This graph shows a large image example
of Computer Science connections when we search
“Java”.
By looking at the graph, we can easily find out
which language skill is the most popular for
recruitment and which skill has strong connections
with others.