Social learning networks: Efficiency optimization for MOOC forums

Brinton, Christopher G.; Buccapatnam, Swapna; Wong, Felix Ming Fai; Chiang, Mung; Poor, H. Vincent

doi:10.1109/infocom.2016.7524579

Cited by 19 publications

(21 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Notice also that β is lowest while α is highest in comp. This observation is consistent with the fact that in humanities courses like comp the discussions in each thread will tend to be longer [2], leading to more new activity notifications, while in engineering courses like ml and algo we would expect learners to more directly answer each other's questions, leading to more explicit replies.…”

Section: Model Analyticssupporting

confidence: 79%

“…where µ u,k denotes the background posting rate of learner u on topic k. Separating the initial posts in threads from future posts in this way enables us to model learners' knowledge seeking (i.e., starting threads) and knowledge disseminating (i.e., posting responses in threads) behavior [2], through the background (µ u,k ) and excitation levels (a u,k ), respectively.…”

Section: Commentmentioning

confidence: 99%

“…where the likelihoods of the initial post and other posts are given by (2) and (5), and the thread text likelihood P(W r |z r = k) is given by the standard LDA model. The threads are then ranked from highest to lowest posting probability.…”

Section: Thread Recommendationmentioning

confidence: 99%

See 2 more Smart Citations

Personalized Thread Recommendation for MOOC Discussion Forums

Lan

Spencer

Chen

et al. 2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Social learning, i.e., students learning from each other through social interactions, has the potential to significantly scale up instruction in online education. In many cases, such as in massive open online courses (MOOCs), social learning is facilitated through discussion forums hosted by course providers. In this paper, we propose a probabilistic model for the process of learners posting on such forums, using point processes. Different from existing works, our method integrates topic modeling of the post text, timescale modeling of the decay in post activity over time, and learner topic interest modeling into a single model, and infers this information from user data. Our method also varies the excitation levels induced by posts according to the thread structure, to reflect typical notification settings in discussion forums. We experimentally validate the proposed model on three real-world MOOC datasets, with the largest one containing up to 6,000 learners making 40,000 posts in 5,000 threads. Results show that our model excels at thread recommendation, achieving significant improvement over a number of baselines, thus showing promise of being able to direct learners to threads that they are interested in more efficiently. Moreover, we demonstrate analytics that our model parameters can provide, such as the timescales of different topic categories in a course.

show abstract

Section: Model Analyticssupporting

confidence: 79%

Section: Commentmentioning

confidence: 99%

See 1 more Smart Citation

Personalized Thread Recommendation for MOOC Discussion Forums

Lan

Spencer

Chen

et al. 2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…The learners may take advantage of these courses, as well as the forums, based on their current needs and their reasons for participating. The increasing interest in the evaluation of MOOC discussion groups, has led researchers to apply a number of tools to assess their quality and efficiency (Brinton et al, ; Lucas, Gunawardena, & Moreira, ) and to tease out key aspects of the interaction that can lead to improvements in online learning environments. To this aim, the current research used a content analysis tool and contributed to further understanding online learners who participate in course forums.…”

Section: Discussionmentioning

confidence: 99%

Active learners’ characterization in MOOC forums and their generated knowledge

Cohen¹,

Shimony²,

Nachmias³

et al. 2018

Brit J Educational Tech

View full text Add to dashboard Cite

This study explores and characterizes learners' participation patterns in MOOC forums, as well as the factors that correlate with learners' participation. Educational data mining and learning analytics methods were used to retrieve and analyze the learners' interpersonal interaction data, which had accumulated in the Coursera log files. The content in the forums was categorized based on Henri's criteria and converted into quantitative values that could be compared and visualized. It was found that only 20% of the learners were collaborating in the forums throughout the entire course and were responsible for 50% of the total posts. A portion of them earned the name "Super Active." The analyses not only demonstrated the volume of activity and its pattern but also revealed the content of the discussions, which helped to highlight the needs and reasons for students' usage of the forums. The content analysis showed intensity in the "Cognitive" and "Discipline" categories. Thus, forum participants benefit from discussions not only socially but disciplinarily and cognitively as well. Furthermore, even though a strong significant correlation was found between the learners' completion status and their activity in the forums, a group of learners, who did not complete the course, was highly active.

show abstract

“…As an example, in Ref. the users' benefit is modeled with general utility functions. Authors evaluate the efficiency of the discussion forums in four MOOC courses, in which they see the potential gains that can be obtained through optimization, proposing for further work to design mechanisms to enforce the optimized networks in practice.…”

Section: Learning Analytics: Research Directionsmentioning

confidence: 99%

Prediction of learning success/failure via pace of events in a social learning network platform

Sousa-Vieira

López-Ardao

Fernández‐Veiga

et al. 2018

Comp Applic In Engineering

View full text Add to dashboard Cite

Earlier studies have shown the power of online learning systems data to develop prediction methods that try to identify successful students patterns of accomplishment and engagement to allow timely pedagogical interventions. Our learning platform, SocialWire, collects a detailed record of the students' activity so, in this paper, we compare and combine the power of different classifiers for success/failure learning prediction, using as inputs some of the features that have measurable correlation with the students' performance. The main conclusion is that, for the courses in study, it is not the type of event/activity initiated by the student what best predicts his/her final grade, but the pace of the events he/she was engaged in.

show abstract

Social learning networks: Efficiency optimization for MOOC forums

Cited by 19 publications

References 13 publications

Personalized Thread Recommendation for MOOC Discussion Forums

Personalized Thread Recommendation for MOOC Discussion Forums

Active learners’ characterization in MOOC forums and their generated knowledge

Prediction of learning success/failure via pace of events in a social learning network platform

Contact Info

Product

Resources

About