JobViz

JobViz

An Interactive Visualization to Help Students Understand Employment and See the "Big Picture"

Understanding how college preparation relates to employment prospects is crucial in positioning a student for successful post-college employment. Yet few tools for job seeking or job market assessment utilize visualizations. Visualizing job postings and helping students gain a better sense of the job market require both the representation of hundreds of majors and thousands of jobs and the display of the intangible relationships between them. To address these challenges, we developed JobViz, an interactive visualization system for university job postings. Combining the intuitiveness of node-link diagrams and the scalability of aggregation-based techniques, JobViz presents a visual overview of the relationships between majors and jobs. The proposed visualization features the simplicity and familiarity of a circular layout, hierarchical edge bundles, and tag clouds and provides both an overview of the database and the capacity for interactive queries to retrieve contextualized details. With this layout, we also tried to remove any inherent bias in the data to enable students to look at and understand all of the data.

The job posting data we used in this project is from Rutgers University Career Services (http://careers.rutgers.edu/) portal. Companies fill out a questionnaire before posting a new job and this information includes: Employer, Job Title, Position Type, Date, Majors, Class Level, City, State, Category, Job Description, and which major they think is most applicable to the job posting. The database contains anywhere from 5000 to 10,000 jobs at any one time.

For information about using or licensing JobViz please refer to the Rutgers Copyright Software Licensing Site.


A DEMO video with brief introduction to the JobViz can be found here.


User Interface

JobViz is a web-based application implemented using D3.js (https://d3js.org/), a JavaScript library for data visualization. The user interface (as shown in Figure 1) consists of four components, including a control panel and an inset of statistical view on the left module, a central module for data visualization display, and a right module for listing and filtering job postings.

Fig. 1. The user interface contains three modules. The left module has (a) a control panel, and (b) an inset of a statistical view. The center module (c) visualizes the job posting database as a circular layout. Majors are on the left side, and career sectors are on the right. A user can select majors or career sectors to see their relationships in a Relation View. Here a Relation View visualizes the links between five selected majors and career fields containing job postings requesting at least one of these majors. (d) The right module lists job postings. In the Tag Cloud View (e), the keywords of job postings that are listed in (d) are displayed as tag cloud. Each term in the tag cloud is clickable and is linked to a list of job postings as in (f). A particular job posting can be identified from (d) or (f).



System Overview

The system has a data processing part and a visualization part. The data visualization contains multiple views that are wellcoordinated: A Database View that is used to provide an overview of all majors, job postings in the database, and their relationships; a Relation View that visualizes the relationships between majors and jobs as spline curves connecting a set of majors with career sectors; and a Tag Cloud View that displays frequent keywords of job postings and links them to their sources visually using color. These views are visually encoded from three data processing steps triggered by user interactions in exploring the visualization: visualization layout creation, major-career links selection, and tag cloud computation. A statistical view is linked to the visualization and shows the distributions of active majors and jobs.

Fig. 2. A system overview of JobViz. The system has a data processing part and a visualization part with three coordinated views: Database View, Relation View, and Tag Cloud View. Additional statistics of majors and jobs can be found in a Statistical View.




Use Cases

1. Getting an Overview



Fig. 3. An overview of the job posting data posted in 2016. The initial Database View in (a) illustrates the distribution of jobs and majorcareer links. The Database View in (b) uses density and thickness of links to show the popularity of majors in the job market. The Tag Cloud Views in (c) and (d) summarize keywords of internships and full-time experienced jobs respectively.


2. Query by College Majors


Fig. 4. Two different Relation Views showing links from the chosen majors to related career sectors. Left: the major group "Social Science Majors" is chosen. Right: the major "Economics" is chosen. The accompanying Statistic Views display the histograms of jobs as bar charts.


3. Query by Career Sectors

Fig. 5. Three different Relation Views showing links from the chosen career sectors to majors listed in those job postings. The “Communication” sector in the left and the “Arts & Entertainment” sector in the center mainly focus on majors of arts, liberal arts, and social science, while the “Science” sector in the right mostly recruits from STEM majors. The accompanying Statistic Views display the distributions of connected majors as pie charts.



Related Publication

[1] Liu, Li, Deborah Silver, and Karen Bemis, "JobViz: Interactive Visualization of Majors & Jobs", Poster Session, IEEE InfoVis 2016, October, 2016.

[2] Liu, Li, D. Silver, and Karen Bemis. "Application Driven-Design: Helping Students Understand Employment and See The “Big Picture”" IEEE Computer Graphics and Applications. (accepted)











 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

 

 

 

 

 

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer