About
This post provides supplemental material and information about the poster "Analyzing MOOC Entries of Professionals on LinkedIn for User Modeling and Personalized MOOC Recommendations: a first look". Available online:
Dataset
| name | number of records | description |
|---|
| users.sql | 5668 | 5668 learner profiles from LinkedIn who have been taken any Coursera MOOCs |
| coruseRecordsV1.sql | 15744 | course records extracted from user profiles |
| eduExperience.sql | 11085 | educational experience of learners |
| workExperience.sql | 32801 | work experience of learners |
| skills.sql | 159291 | skills of learners |
Descriptive statistics: the dataset is about analyzed MOOC learner profiles from LinkedIn, which consists of 15,744 MOOC entries from 5,668 professionals. Each professional took 3 courses on average with the majority of learners (87%) having less than or equal to 5 MOOCs. Interestingly, the learner with the largest number of MOOCs had 114 of them. The distribution of genders and degrees of learners is as below:
If we assume that course entries in LinkedIn are courses that have been completed by users, the distribution of degrees are similar to the study [1] which provides the distribution of learners who completed their course.
Verified certifications: Instead of just taking MOOCs on Coursera and getting statements of accomplishment, learners can also purchase verified certifications for some courses that meet certain criteria. A verified certification provides proof that learners have completed their online courses. In such cases, varied certifications can also be added to LinkedIn parallels with their varied serial numbers. We found that around 26% of certifications in our collected profiles are verified while 74% of the certifications are unverified.
Course tracks. We found that course tracks can be identified by exploring learning activities of users in the OSN. Formally, we can define a course track as a set of courses that were taken together more than n times where n is a threshold. The course relationships can be represented by weighted undirected networks like in the figure below.
Nodes denote courses and the ties among courses denote the frequency of two courses taken together. In this context, a course track is a clique (or complete graph that has an edge joining each pair of nodes) within the course relationships network, with the weight of each tie in the clique is higher than the threshold n. Course tracks can be constructed based on the cliques within the course relationships network. As one might expect, the higher of the value n, the stronger the relationships a course track must hold with less number of cliques meeting the criteria. Indeed, 60 maximal cliques (a clique in maximal if it cannot be extended to a larger clique) can be found with a threshold of 10 while 16 maximal cliques can be found with a threshold of 20.
We evaluated these tracks and found that two of the course tracks provided by Coursera can be identified in those cliques through this approach. A course track, called a specialization in Coursera, is a targeted sequence of courses from an institution taken together to earn a specialization certificate. The first course track from Coursera is a specialization of "Data Science" which consists of 9 courses from Johns Hopkins University, and the second one is a specialization of "Business Foundations" provided by the University of Pennsylvania. In practice, these ground truth course tracks can also be used for identifying the threshold n, which is the highest value that does not break the ground truth course tracks. In our case, 27 maximal cliques can be found including the two golden truth course tracks with the value of 13 for the threshold. Interestingly, when we look at the maximal clique that contains the "Data Science" course track (Figure 2), we found that "Machine Learning", "Introduction to Data Science" and "Computing for Data Analysis" are also being taken frequently with 9 courses in the "Data Science" course track in practice. This indicates that new course tracks can be constructed on top of existing tracks by exploring learning activities of users from the OSN.
