A Multigranularity Text Driven Named Entity Recognition CGAN Model for Traditional Chinese Medicine Literatures.

Recognition of Traditional Chinese Medicine (TCM) entities from different types of literature is challenging research, which is the foundation for extracting a large amount of TCM knowledge existing in unstructured texts into structured formats. The lack of large-scale annotated data makes unsatisfactory application of conventional deep learning models in TCM text knowledge extraction. Some other unsupervised methods rely on other auxiliary data, such as domain dictionaries. We propose a multigranularity text-driven NER model based on Conditional Generation Adversarial Network (MT-CGAN) to implement TCM NER with small-scale annotated corpus. In the model, a multigranularity text features encoder (MTFE) is designed to extract rich semantic and grammatical information from multiple dimensions of TCM texts. By differentiating the conditional constraints of the generator and discriminator of MT-CGAN, the synchronization between the generated tag labs and the named entities is guaranteed. Furthermore, seeds of different TCM text types are introduced into our model to improve the precision of NER. We compare our method with other baseline methods to illustrate the effectiveness of our method on 4 kinds of gold-standard datasets. The experiment results show that the standard precision, recall, and F1 score of our method are higher than the state-of-the-art methods by 0.24∼8.97%, 0.89∼12.74%, and 0.01∼10.84%. MT-CGAN is able to extract entities from different types of TCM literature effectively. Our experimental results indicate that the proposed approach has a clear advantage in processing TCM texts with more entity types, higher sparsity, less regular features, and a small-scale corpus.

Excessive Nitrogen Fertilization Favors the Colonization, Survival, and Development of Sogatella furcifera via Bottom-Up Effects.

Fertilization can trigger bottom-up effects on crop plant-insect pest interactions. The intensive use of nitrogen fertilizer has been a common practice in rice production, while the yield has long been challenged by the white-backed planthopper, Sogatella furcifera (Horváth). High nitrogen fertilization can facilitate S. furcifera infestation, however, how nitrogen fertilizer leads to high S. furcifera infestation and the nutritional interactions between rice and S. furcifera are poorly understood. Here, we evaluated the effects of various levels of nitrogen fertilizer application (0-350 kg/ha) on rice, and subsequently on S. furcifera performance. We found that higher nitrogen fertilizer application: (1) increases the preference of infestation behaviors (feeding and oviposition), (2) extends infestation time (adult lifespan), and (3) shortens generation reproduction time (nymph, pre-oviposition, and egg period), which explain the high S. furcifera infestation ratio on rice paddies under high nitrogen conditions. Moreover, high nitrogen fertilizer application increased all tested rice physical indexes (plant height, leaf area, and leaf width) and physiological indexes (chlorophyll content, water content, dry matter mass, and soluble protein content), except for leaf thickness, which was reduced. Correlation analysis indicated that the specific rice physical and/or physiological indexes were conducive to the increased infestation behavior preference, extended infestation time, and shortened generation reproduction time of S. furcifera. The results suggested that nitrogen fertilizer triggers bottom-up effects on rice and increases S. furcifera populations. The present study provides an insight into how excess nitrogen fertilization shapes rice-planthopper interactions and the consequent positive effect on S. furcifera infestation.

Traditional Chinese medicine symptom normalization approach leveraging hierarchical semantic information and text matching with attention mechanism.

Traditional Chinese medicine (TCM) symptom normalization is difficult because the challenges of the symptoms having different literal descriptions, one-to-many symptom descriptions and different symptoms sharing a similar literal description. We propose a novel two-step approach utilizing hierarchical semantic information that represents the functional characteristics of symptoms and develop a text matching model that integrates hierarchical semantic information with an attention mechanism to solve these problems. In this study, we constructed a symptom normalization dataset and a TCM normalization symptom dictionary containing normalization symptom words, and assigned symptoms into 24 classes of functional characteristics. First, we built a multi-label text classifier to isolate the hierarchical semantic information from each symptom description and count the corresponding normalization symptoms and filter the candidate set. Then we designed a text matching model of mixed multi-granularity language features with an attention mechanism that utilizes the hierarchical semantic information to calculate the matching score between the symptom description and the normalization symptom words. We compared our approach with other baselines on real-world data. Our approach gives the best performance with a Hit@ 1, 3, and 10 of 0.821, 0.953, and 0.993, respectively, and a MeanRank of 1.596, thus outperforming significantly regarding the symptom normalization task. We developed an approach for the TCM symptom normalization task and demonstrated its superior performance compared with other baselines, indicating the promise of this research direction.

Influence of Supplemental Protein on the Life Expectancy and Reproduction of the Chinese Citrus Fruit Fly, Bactrocera minax (Enderlein) (Tetradacus minax) (Diptera: Tephritidae).

Bactrocera minax (Enderlein) (Diptera: Tephritidae) is a major citrus pest in China, whose artificial rearing technology of the adult is not well documented to date. In this study, we tried to determine if supplementing proteins to the adult diet could result in the enhancement of some fitness parameters of B. minax. Four feeds with varying protein source were provided as F0 (water), F1 (sucrose), F2 (sucrose + yeast), and F3 (sucrose + peptone). F0 and F1 being the control, F2 and F3 were protein food types. The results showed that adults fed by F2 and F3 lived longer with 40.1 d and 32.8 d, respectively, had reduced death rates (death peaks were delayed for 5.6 d and 4.1 d, respectively), increased mating frequencies (8.1 and 5.3 per females, 4.7 and 7.3 per males, respectively), and longer mating durations (with 42 d and 34 d). In addition, females recorded an increased adult ovary development, more egg load (with 94.8 and 77.3 brood eggs per ovary) and to greater oviposition rates of 63.2 eggs/female and 19.3 eggs/female. Based on our results, protein supplements enhanced B. minax survival, mating, and fecundity. This study does not only provide basic knowledge to implement artificial rearing of B. minax, but also deepens our understanding on its physiology that could be used to enhance the management of the pest.

Distributed Task Offloading in Heterogeneous Vehicular Crowd Sensing.

The ability of road vehicles to efficiently execute different sensing tasks varies because of the heterogeneity in their sensing ability and trajectories. Therefore, the data collection sensing task, which requires tempo-spatial sensing data, becomes a serious problem in vehicular sensing systems, particularly those with limited sensing capabilities. A utility-based sensing task decomposition and offloading algorithm is proposed in this paper. The utility function for a task executed by a certain vehicle is built according to the mobility traces and sensing interfaces of the vehicle, as well as the sensing data type and tempo-spatial coverage requirements of the sensing task. Then, the sensing tasks are decomposed and offloaded to neighboring vehicles according to the utilities of the neighboring vehicles to the decomposed sensing tasks. Real trace-driven simulation shows that the proposed task offloading is able to collect much more comprehensive and uniformly distributed sensing data than other algorithms.