In a user engagement prediction paradigm we compare all features against semi-automatic (cross-task), and automatic (task-specific) feature selection methods. From this, an effective combination of features is outlined for approaches aiming to automatically predict several user engagement indicators. Furthermore, the extracted time-series features reveal significant (p < 0.05) correlations for each dimension, such as, count below signal mean (arousal), number of peaks (valence), and absolute energy (trustworthiness).

Our results demonstrate that smaller boundary ranges and fluctuations for arousal lead to an increase in user engagement.

In doing so, we identify the positive and negative influences which single features have, as well as interpretable patterns in each dimension which relate to user engagement. We investigate features extracted from these signals against various user engagement indicators including views, like/dislike ratio, as well as the sentiment of comments. 600 h of dimensional arousal, valence and trustworthiness annotations. In this contribution, we utilize a large dataset of YouTube review videos which includes ca. This limited understanding is partly due to a scarcity in emotionally annotated data and the varied modalities which express user engagement online. However, the causal relationship between these signals and online user engagement is not well understood. Portraying emotion and trustworthiness is known to increase the appeal of video content. Experiments were carried out on question-answers datasets (Quora (QQP), TREC QA, and Yahoo Answers) as well as e-books on various subjects and across different grades.The model outperforms previous research by a large margin. This effort also provides advanced video indexing methods for each recommended video, allowing videos with “Table of Contents” and “Phrase Cloud” features to make videos more consumable. The model identifies the key topics of the e-book, after which sBERT is used to assess the similarity between the question-answers. We construct SiameseBERT (Bidirectional Encoder Representations from Transformers) network which provides the semantically relevant phrase embeddings. To address this issue, we propose a novel sBERT-LDA model, which augments the e-books with recommended question-answers and videos. As a result, it fails to find the semantically cohesive, and meaningful topics. The major flaw of this LDA is that it can’t capture the semantic knowledge in the documents. The Latent Dirichlet Allocation (LDA) topic model helps us to find the key topic of the e-book. On the other hand, e-book is primary source of knowledge for students. However, there is a redundancy issue which results in inadequate search results for a given question.

The community question-answering (CQA) websites such as Quora¹, and Reddit², and YouTube provides a significant resource to the students.

Video media meta data tags