We present an event extraction framework to detect event mentions and extract events from the document-level financial news. Up to now, methods based on supervised learning paradigm gain the highest performance in public datasets (such as ACE 2005 1 , KBP 2015 2). These methods heavily depend on the manually labeled training data. However, in particular areas, such as financial, medical and judicial domains, there is no enough labeled data due to the high cost of data labeling process. Moreover, most of the current methods focus on extracting events from one sentence, but an event is usually expressed by multiple sentences in one document. To solve these problems, we propose a Document-level Chinese Financial Event Extraction (DCFEE) system which can automatically generate a large scaled labeled data and extract events from the whole document. Experimental results demonstrate the effectiveness of it.

Word embeddings and convolutional neural networks (CNN) have attracted extensive attention in various classification tasks for Twitter, e.g. sentiment classification. However, the effect of the configuration used to train and generate the word embeddings on the classification performance has not been studied in the existing literature. In this paper, using a Twitter election classification task that aims to detect election-related tweets, we investigate the impact of the background dataset used to train the embedding models, the context window size and the dimensionality of word embeddings on the classification performance. By comparing the classification results of two word embedding models, which are trained using different background corpora (e.g. Wikipedia articles and Twitter microposts), we show that the background data type should align with the Twitter classification dataset to achieve a better performance. Moreover, by evaluating the results of word embeddings models trained using various context window sizes and dimensionalities, we found that large context window and dimension sizes are preferable to improve the performance. Our experimental results also show that using word embeddings and CNN leads to statistically significant improvements over various baselines such as random, SVM with TF-IDF and SVM with word embeddings.

Shale hydration and swelling are disadvantageous for well drilling, especially when using water-based drilling fluids. In this work, the ionic liquid 1-vinyl-3-ethylimidazolium bromide (VeiBr) monomer and its corresponding homopolymers (PV) were innovatively used as shale hydration inhibitors. Both composites of sodium bentonite (Na-BT) with VeiBr and PV (hereafter denoted as Na-BT/VeiBr and Na-BT/PV composites) exhibited excellent temperature stability up to 300 °C, showing potential application in high-temperature well drilling. The inhibiting performance was evaluated by measuring the linear swelling height, rheological property of Na-BT aqueous solutions, and recovery percentage of shale cuttings after hot rolling. Results indicated that VeiBr monomer and PV polymer displayed better inhibition performance than inorganic KCl and organic quaternary amine 2,3-epoxypropyltrimethylammonium chloride in all tests. In addition, PV was even better than VeiBr. The underlying mechanism was analyzed by measuring the interlayer distance through X-ray diffraction, observing the aggregation through scanning electron microscopy, and determining the ζ potential and particle size distribution. The monomer exerted its effect mainly by decreasing the interlayer spacing, whereas the polymer increased the viscosity, encapsulated Na-BT particles, prevented the exfoliation of Na-BT, and decreased the interlayer spacing depending upon the molecular weight. This study can serve as a basis for using ionic liquids in the design of permanent shale inhibitors for drilling fluids.

In this work, a new cost-effective, rapid and simple method for the preparation of stable silver nanoparticles (AgNPs) was developed, which can be completed within 15 minutes at room temperature by oxidizing the reductants in pear juice with AgNO3. Compared with the most used citrate-capped AgNPs, the as-prepared AgNPs showed high stability, good biocompatibility and enhanced antibacterial activity. Based on the formation of Ag-S covalent bonds between cysteine and AgNPs as well as the electrostatic interaction of COO(-) and NH4(+) between cysteine molecules, which selectively lead to the aggregation of the as-prepared AgNPs and give a specific yellow-to-red colour change, a new selective colorimetric method for detection of cysteine was proposed with the as-prepared AgNPs by coupling the decrease of the characteristic localized surface plasmon resonance (LSPR) absorption at 406 nm of the as-prepared AgNPs and the increase of the new aggregation-induced band at 530 nm. The ratio of the absorbance at 530 nm to 406 nm (A530/A406) was found to be linearly dependent on the cysteine concentrations in the range of 5.0 × 10(-7) to 1.0 × 10(-5) M with a limit of detection of 6.8 × 10(-8) M.

Previously, our group developed a vinylimidazolium-based ionic liquid (IL) as an excellent shale hydration inhibitor for water-based drilling fluids (WBDFs). Herein, several ILs with different alkyl-chain lengths on the vinylimidazolium group were successfully synthesized by adjusting the cation composition to study their influence on inhibition performance. The results indicated that the IL with an ethyl group (C2) showed the strongest inhibitory effects for bentonite swelling, shalecutting dispersion and rheological properties of bentonite suspension. Furthermore, the IL inhibition performance decreased with increasing alkyl-chain length. Accordingly, we concluded that as alkyl-chain length increased, the IL molecular volume increased, while the IL hydrophilicity and solubility decreased; minimizing the interlayer space and decreasing the water activity became more difficult, thus decreasing their inhibiting performance. Simultaneously, the reduction in inhibition performance has little relationship with the ability to suppress the double electron layers. All these findings can serve as a basis for designing ILs for high-performance shale hydration inhibition in WBDFs.