The impact of automatic text translation on classification of online discussions for social and cognitive presences

Can Multi-Label Classifiers Help Identify Subjectivity? A Deep Learning Approach to Classifying Cognitive Presence in MOOCs

“…The outcomes of the experiments in the paper also indicate that the performance of the automatic classifiers might have some relevance to the distribution of the cognitive presence phases. The classifiers in the study reached the similar level of prediction performance to the classifiers in most of the previous studies in the literature (Barbosa et al, 2021;Farrow et al, 2019;Hu et al, 2021b;Kovanović et al, 2016;Waters et al, 2015). We found the cognitive presence phases in the training set (AgreementSet) of this study obtained very similar distribution to the data sets in the previous studies, where the highest proportion located in Exploration and Integration, and the lowest in the Other and Resolution.…”

“…Neto et al's (2018) study reached the accuracy of 83% and Cohen's κ of 0.72 in the Portuguese discussion data, and Barbosa et al' (2020) study achieved the accuracy of 67% and Cohen's κ of 0.32 in cross-language (English & Portuguese) discussion data. Barbosa et al (2021) then extended the RF classifier using automatic text translation method to analyse both cognitive and social presence in the cross-language discussion messages and obtained similar performance results to the previous studies. Neto et al (2021) also applied the RF classifiers to classify the discussion messages from two discipline courses (biology & technology), reaching Cohen's κ of 0.55 in the experiments of using combined data sets and the Cohen's κ of below 0.4 in the crossdiscipline tests.…”

“…Only the single-label classification approaches have been attempted in the literature of automatic classification of cognitive presence in online discussions. The RF approach and its extensive use was regarded as the state-ofthe-art single-label classifier for analysing cognitive presence in online discussions (Barbosa et al, 2021;Farrow et al, 2019;Kovanović et al, 2016;Neto et al, 2018).…”

Analysis and Application of the Business English Translation Query and Decision Model with Big Data Corpus

“…On the FNR side, compared with the two best performing models A and B in all comparison models, in this paper, the model decreased by 6.53% (5.05%⟶4.72%) and 10.78% (5.29%⟶4.72%), respectively. On the FPR, compared with the two best performing models A and C in all comparison models, the proposed model decreases by 10.29% (4.86%→4.36%) and 13.32% (5.03%→4.36%) Input: Initialize weights W xi and W hi , encoder features (1) Using formula (7) to compute the forgetting gate feature vector ft;…”

“…Among them, the corpus is the carrier of knowledge representation and storage. At present, the corpus mainly expresses the semantic relationship between the sentence to be queried and empirical knowledge by using triples [7]. e query and decision-making task of the corpus is to identify the natural language which included entities, entity relationships, and entity types.…”