Automatic Grading System Using Sentence-BERT Network

Liu

2023

Sci Rep

In recent years, short videos have become an increasingly vital source of information. To compete for users’ attention, short video platforms have been overusing algorithmic technology, making the group polarization intensify, which is likely to push users into the homogeneous “echo chamber”. However, echo chambers can contribute to the spread of misleading information, false news, or rumors, which have negative social impacts. Therefore, it is necessary to explore echo chamber effects in short video platforms. Moreover, the communication paradigms between users and feed algorithms greatly vary across short video platforms. This paper investigated echo chamber effects of three popular short video platforms (Douyin, TikTok, and Bilibili) using social network analysis and explored how user features influenced the generation of echo chambers. We quantified echo chamber effects through two primary ingredients: selective exposure and homophily, in both platform and topic dimensions. Our analyses indicate that the gathering of users into homogeneous groups dominates online interactions on Douyin and Bilibili. We performed performance comparison of echo chamber effects and found that echo chamber members tend to display themselves to attract the attention of their peers and that cultural differences can prevent the development of echo chambers. Our findings are of great value in designing targeted management strategies to prevent the spread of misleading information, false news, or rumors.

Section: Ec Effect Measurementmentioning

confidence: 99%

Echo chamber effects on short video platforms

Liu

2023

Sci Rep

“…Though ASAG is crucial, implementing these expensive models may pose challenges. In this research, we explore some other Sentence-Bidirectional Encoder Representations (SBERT) models as mentioned in [15], and then propose a simpler model by fine-tuning certain hyperparameters to optimize the ASAG performance.…”

Section: Introductionmentioning

confidence: 99%

Combining Balancing Dataset and SentenceTransformers to Improve Short Answer Grading Performance

Wijanto,

Yong

2024

Applied Sciences

Short-answer questions can encourage students to express their understanding. However, these answers can vary widely, leading to subjective assessments. Automatic short answer grading (ASAG) has become an important field of research. Recent studies have demonstrated a good performance using computationally expensive models. Additionally, available datasets are often unbalanced in terms of quantity. This research attempts to combine a simpler SentenceTransformers model with a balanced dataset, using prompt engineering in GPT to generate new sentences. Our recommended model also tries to fine-tune several hyperparameters to achieve optimal results. The research results show that the relatively small-sized all-distilroberta-v1 model can achieve a Pearson correlation value of 0.9586. The RMSE, F1-score, and accuracy score also provide better performances. This model is combined with the fine-tuning of hyperparameters, such as the use of gradient checkpointing, the split-size ratio for testing and training data, and the pre-processing steps. The best result is obtained when the new generated dataset from the GPT data augmentation is implemented. The newly generated dataset from GPT data augmentation achieves a cosine similarity score of 0.8 for the correct category. When applied to other datasets, our proposed method also shows an improved performance. Therefore, we conclude that a relatively small-sized model combined with the fine-tuning of the appropriate hyperparameters and a balanced dataset can provide performance results that surpass other models that require larger resources and are computationally expensive.

Int. J. Adv. Corp. Learn.

“…For instance, researchers used deep learning techniques for AQG such as BERT [2], T5 transformer language model [8], GPT-2, and GPT-3 language model [11]. Moreover, a text-based similarity measure such as sentence-BERT (SBERT) [12] was used for automatic answer assessment (AAA) or grading [10]. The paradigm of AQG and AAA can make teachers more efficient.…”

Section: Introductionmentioning

confidence: 99%

VideoDL: Video-Based Digital Learning Framework Using AI Question Generation and Answer Assessment

Forkan

Kang

Jayaraman

et al. 2023

Assessing learners’ understanding and competency in video-based digital learning is time-consuming and very difficult for educators, as it requires the generation of accurate and valid questions from pre-recorded learning videos. This paper demonstrates VideoDL, a video-based learning framework powered by Artificial Intelligence (AI) that supports automatic question generation and answer assessment from videos. VideoDL comprises of various AI algorithms, and an interactive web-based user interface (UI) developed using the principles of human-centred design. Our empirical evaluation using real-world videos from multiple domains demonstrates the effectiveness of VideoDL.