A novel video recommendation system for algebra: An effectiveness evaluation study

Leite, Walter L.; Roy, Samrat Guha; Chakraborty, N.; Michailidis, George; Huggins-Manley, Anne Corinne; D’Mello, Sidney K.; Faradonbeh, Mohamad Kazem Shirani; Jensen, Emily; Kuang, Huan; Jing, Zeyuan

doi:10.1145/3506860.3506906

Cited by 12 publications

(13 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The higher the CTR, the longer the session length, and the better the recommendation quality. The compiler average causal effect (CACE) evaluator was employed by Leite et al (2022) to test the impact of recommendations offered to the treatment group. Breitfuss et al (2021) tested their knowledge graphbased recommendation strategy by various metrics, including Sparsity impact, the granularity of emotions, extensibility, recommendation quality, and additional characteristics.…”

Section: Evaluation Methodsmentioning

confidence: 99%

“…Deep learning (DL) has gradually penetrated the field of affective computing and promoted the development of video recommendations. Deep learning-based models applied in AVRS in recent years include reinforcement learning (RL) (Leite et al, 2022), convolutional neural network (CNN) (Zhu et al, 2019), long short-term memory (LSTM) (Cao et al, 2022), multilayer perception (Ðor dević Čegar et al, 2020) (MLP), deep hybrid models (DHM) (Mishra et al, 2020), etc. The evolution of AVRS with different algorithms and databases is illustrated in Figure 2.…”

Section: The State-of-the-art Affective Video Recommendation Algorith...mentioning

confidence: 99%

“…The core idea of adopting RL in AVRS is that DL can continuously and dynamically learn strategies through the real-time state changes caused by the impact of users on the surrounding environment to maximize the cumulative reward. In Leite et al (2022), they discussed the role of deep reinforcement learning (DRL) in video recommendation when used in a virtual learning environment. They also considered two different student groups, i.e., common effect and high effect.…”

Section: Reinforcement Learning (Rl)mentioning

confidence: 99%

“…Facial expressions/features Soleymani and Pantic, 2012;Zhao et al, 2013;Boughrara et al, 2016;Choi et al, 2016;Kaklauskas et al, 2016;Mahata et al, 2017;Diaz et al, 2018;Fonnegra, 2018;Hewitt and Gunes, 2018;Kaklauskas et al, 2018;Bohlin et al, 2019;Soni et al, 2019;De Pessemier et al, 2020;Mishra et al, 2020;Leite et al, 2022 Skin-estimated pulse/heart rate Dabas et al, 2018;Diaz et al, 2018;Shu et al, 2018a;Bohlin et al, 2019;Soni et al, 2019;Ðor dević Čegar et al, 2020Mood Winoto and Tang, 2010EDA Ðor dević Čegar et al, 2020BA Alhagry, 2017Liu et al, 2017;Kwon et al, 2018;Ogawa et al, 2018;Yang et al, 2019;Ðor dević Čegar et al, 2020 User interactions Niu et al, 2013;Niu et al, 2016GSR Kwon et al, 2018 Body gestures Hassib et al, 2017 Perceived connotative properties Martha and Larson, 2013;Zhang and Zhang, 2017 Movie reviews/comments/web recordings Mulholland et al, 2017;Yenter, 2017;Tripathi et al, 2019;Krishnamurthy, 2020;Pan et al, 2020;Breitfuss et al...…”

Section: Audience Responses Publicationsmentioning

confidence: 99%

See 3 more Smart Citations

Affective video recommender systems: A survey

Wang

Zhao

2022

Front. Neurosci.

View full text Add to dashboard Cite

Traditional video recommendation provides the viewers with customized media content according to their historical records (e.g., ratings, reviews). However, such systems tend to generate terrible results if the data is insufficient, which leads to a cold-start problem. An affective video recommender system (AVRS) is a multidiscipline and multimodal human-robot interaction (HRI) system, and it incorporates physical, physiological, neuroscience, and computer science subjects and multimedia resources, including text, audio, and video. As a promising research domain, AVRS employs advanced affective analysis technologies in video resources; therefore, it can solve the cold-start problem. In AVRS, the viewers’ emotional responses can be obtained from various techniques, including physical signals (e.g., facial expression, gestures, and speech) and internal signals (e.g., physiological signals). The changes in these signals can be detected when the viewers face specific situations. The physiological signals are a response to central and autonomic nervous systems and are mostly involuntarily activated, which cannot be easily controlled. Therefore, it is suitable for reliable emotion analysis. The physical signals can be recorded by a webcam or recorder. In contrast, the physiological signals can be collected by various equipment, e.g., psychophysiological heart rate (HR) signals calculated by echocardiogram (ECG), electro-dermal activity (EDA), and brain activity (GA) from electroencephalography (EEG) signals, skin conductance response (SCR) by a galvanic skin response (GSR), and photoplethysmography (PPG) estimating users’ pulse. This survey aims to provide a comprehensive overview of the AVRS domain. To analyze the recent efforts in the field of affective video recommendation, we collected 92 relevant published articles from Google Scholar and summarized the articles and their key findings. In this survey, we feature these articles concerning AVRS from different perspectives, including various traditional recommendation algorithms and advanced deep learning-based algorithms, the commonly used affective video recommendation databases, audience response categories, and evaluation methods. Finally, we conclude the challenge of AVRS and provide the potential future research directions.

show abstract

Section: Evaluation Methodsmentioning

confidence: 99%

Section: The State-of-the-art Affective Video Recommendation Algorith...mentioning

confidence: 99%

Section: Reinforcement Learning (Rl)mentioning

confidence: 99%

Section: Audience Responses Publicationsmentioning

confidence: 99%

See 2 more Smart Citations

Affective video recommender systems: A survey

Wang

Zhao

2022

Front. Neurosci.

View full text Add to dashboard Cite

show abstract

“…The authors found that recommender system techniques such as matrix factorization and collaborative filtering outperformed the traditional regression methods in predicting student performance. Recently, more advanced algorithms and approaches such as reinforcement learning [11,19] have been proposed to improve the capacity, architectural flexibility, and performance accuracy of the recommender systems adopted within educational contexts.…”

Section: Recommender Systems For Educational Assessmentsmentioning

confidence: 99%

Learning Analytics and Computerized Formative Assessments: An Application of Dijkstra’s Shortest Path Algorithm for Personalized Test Scheduling

2022

View full text Add to dashboard Cite

The use of computerized formative assessments in K–12 classrooms has yielded valuable data that can be utilized by learning analytics (LA) systems to produce actionable insights for teachers and other school-based professionals. For example, LA systems utilizing computerized formative assessments can be used for monitoring students’ progress in reading and identifying struggling readers. Using such LA systems, teachers can also determine whether progress is adequate as the student works towards their instructional goal. However, due to the lack of guidelines on the timing, number, and frequency of computerized formative assessments, teachers often follow a one-size-fits-all approach by testing all students together on pre-determined dates. This approach leads to a rigid test scheduling that ignores the pace at which students improve their reading skills. In some cases, the consequence is testing that yields little to no useful data, while increasing the amount of instructional time that students miss. In this study, we propose an intelligent recommender system (IRS) based on Dijkstra’s shortest path algorithm that can produce an optimal assessment schedule for each student based on their reading progress throughout the school year. We demonstrated the feasibility of the IRS using real data from a large sample of students in grade two (n = 668,324) and grade four (n = 727,147) who participated in a series of computerized reading assessments. Also, we conducted a Monte Carlo simulation study to evaluate the performance of the IRS in the presence of unusual growth trajectories in reading (e.g., negative growth, no growth, and plateau). Our results showed that the IRS could reduce the number of test administrations required at both grade levels by eliminating test administrations in which students’ reading growth did not change substantially. In addition, the simulation results indicated that the IRS could yield robust results with meaningful recommendations under relatively extreme growth trajectories. Implications for the use of recommender systems in K–12 education and recommendations for future research are discussed.

show abstract