Search

Learners in a massive open online course often express feelings, exchange ideas and seek help by posting questions in discussion forums. Due to the very high learner-to-instructor ratios, it is unrealistic to expect instructors to adequately track the forums, find all of the issues that need resolution and understand their urgency and sentiment. In this paper, considering the biases among different courses, we propose a transfer learning framework based on a convolutional neural network and a long short-term memory model, called ConvL, to automatically identify whether a post expresses confusion, determine the urgency and classify the polarity of the sentiment. First, we learn the feature representation for each word by considering the local contextual feature via the convolution operation. Second, we learn the post representation from the features extracted through the convolution operation via the LSTM model, which considers the long-term temporal semantic relationships of features. Third, we investigate the possibility of transferring parameters from a model trained on one course to another course and the subsequent fine-tuning. Experiments on three real-world MOOC courses confirm the effectiveness of our framework. This work suggests that our model can potentially significantly increase the effectiveness of monitoring MOOC forums in real time.

The literature [-5]contains several reports evaluating the abilities of deep neural networks in text transfer learning. To our knowledge, however, there have been few efforts to fully realize the potential of deep neural networks in cross-domain product review sentiment classification. In this paper, we propose a two-layer convolutional neural network (CNN) for cross-domain product review sentiment classification (LM-CNN-LB). Transfer learning research into product review sentiment classification based on deep neural networks has been limited by the lack of a large-scale corpus; we sought to remedy this problem using a large-scale auxiliary cross-domain dataset collected from Amazon product reviews. Our proposed framework exhibits the dramatic transferability of deep neural networks for cross-domain product review sentiment classification and achieves state-of-the-art performance. The framework also outperforms complex engineered features used with a non-deep neural network method. The experiments demonstrate that introducing large-scale data from similar domains is an effective way to resolve the lack of training data. The LM-CNN-LB trained on the multi-source related domain dataset outperformed the one trained on a single similar domain.

Medical images are valuable for clinical diagnosis and decision making. Image modality is an important primary step, as it is capable of aiding clinicians to access required medical image in retrieval systems. Traditional methods of modality classification are dependent on the choice of hand-crafted features and demand a clear awareness of prior domain knowledge. The feature learning approach may detect efficiently visual characteristics of different modalities, but it is limited to the number of training datasets. To overcome the absence of labeled data, on the one hand, we take deep convolutional neural networks (VGGNet, ResNet) with different depths pre-trained on ImageNet, fix most of the earlier layers to reserve generic features of natural images, and only train their higher-level portion on ImageCLEF to learn domain-specific features of medical figures. Then, we train from scratch deep CNNs with only six weight layers to capture more domain-specific features. On the other hand, we employ two data augmentation methods to help CNNs to give the full scope to their potential characterizing image modality features. The final prediction is given by our voting system based on the outputs of three CNNs. After evaluating our proposed model on the subfigure classification task in ImageCLEF2015 and ImageCLEF2016, we obtain new, state-of-the-art results—76.87% in ImageCLEF2015 and 87.37% in ImageCLEF2016—which imply that CNNs, based on our proposed transfer learning methods and data augmentation skills, can identify more efficiently modalities of medical images.

Rocky desertification is a major ecological problem of land degradation in karst areas. Its high soil calcium (Ca2+) content has become an important environmental factor which can affect the restoration of vegetation in such rocky desertification areas. Consequently, the screening of plant species, which can adapt to soils high Ca2+ environment, is a critical step for vegetation restoration. However, the Ca2+ dynamics of plants and soil are not well understood. In this study, three different grades of rocky desertification samples areas (LRD, light rocky desertification; MRD, moderate rocky desertification; IRD, intense rocky desertification) were selected in karst areas of southwestern Hunan, China. Each grade of these samples areas had 3 sample plots in different slop positions, each of which had 4 small quadrats (1 in rocky side areas, 3 in non-rocky side areas). We measured the Ca2+ content of leaves, branches and roots from 41 plant species, as well as soil total Ca2+ (TCa) and exchange Ca2+ (ECa) at depths of 0-15, 15-30 and 30-45 cm under each small quadrat. The results showed that the soil Ca2+ content in rocky side areas was significantly higher than that in non-rocky side areas (p < 0.05). The mean soil TCa and ECa content increased gradually along with the grade of rocky desertification, in the order IRD > MRD > LRD. For all plant functional groups, the plant Ca2+ content of aboveground parts was significantly higher than that of the underground parts (p < 0.05). The soil ECa content had significant effects on plant Ca2+ content of the underground parts, but had no significant effects on plant Ca2+ content of the aboveground parts. According to the differences in Ca2+ content between the aboveground and underground parts of 17 dominant species (important value, IV > 1) and their correlations with soil ECa content, these 17 species can be divided into three categories: Ca-indifferent plants, high-Ca plants and low-Ca plants. Our results can provide a vital theoretical basis and practical guide for vegetation restoration and ecosystem reconstruction in rocky desertification areas. [ABSTRACT FROM AUTHOR]

Cross-domain sentiment classification trains robust classifiers across domains with the help of source domain labeled data. Sentiment is expressed differently in different domains. Sentiment terms that occur in a source domain may not appear in a different target domain. Such feature mismatches hinder cross-domain sentiment classification. Previous studies have addressed this problem by constructing a common feature representation or subspace; however, they have not considered semantic correlations between features. In this paper, we propose a cross-domain semantic correlation auto-correspondence method (CSCW) by capturing similar semantic features from different domains. First, our method uses sentiment-invariance words as features by considering their properties as strong sentiment indicators and their invariance across domains. Second, we extracted the top-N pivot features using a common frequency among source and target domains. These pivot features can then be employed to find semantically similar sentiment features from both domains. Third, for each pivot feature, by calculating the semantic similarity between non-pivot features and pivot features from either domain with the help of Word2Vec, we construct similar-pivot feature pairs that express similar sentiments but in different representations in either domain. Finally, we transform these pairs to align similar sentiment feature representations. This process avoids feature mismatches and reduces sentiment discrepancies between domains. The experimental results from testing on 12 source-target domain pairs of Amazon product reviews demonstrate that our method significantly outperforms previous approaches in sentiment classification. [ABSTRACT FROM AUTHOR]

Sedum aizoon is an edible medicinal plant with various pharmacological activities distributed widely in China. The flavonoids of S. aizoon have been shown to possess a variety of important biological activities. In this work, a microwave-assisted extraction (MAE) method was employed for the efficient extraction of flavonoids from S. aizoon leaves. The influences of MAE conditions including temperature, extraction time, solvent to solid ratio, and ethanol concentration on the yield of flavonoids were systematically investigated by single-factor experiments and response surface methodology (RSM) experimental design. Under the optimal MAE conditions of extraction time 20 min, extraction temperature 57°C, solvent to solid ratio 20mL/g, and ethanol concentration 80.6%, the flavonoids yield reached 24.87 ± 0.26 mg/g and was much higher than that of conventional Soxhlet extraction (CSE) method (18.67 ± 0.35 mg/g). The obtained flavonoids exhibited strong antioxidant activities with IC50 value of 0.315 mg/mL in the DPPH radical-scavenging experients. The results indicated that MAE was a simple and efficient technology for the extraction of flavonoids from S. aizoon leaves and the flavonoids may serve as new potential natural antioxidant functional food ingredients and additives. [ABSTRACT FROM AUTHOR]