Improving Accuracy using The ASERLU layer in CNN-BiLSTM Architecture on Sentiment Analysis

There have been 350,000 tweets generated by the interaction of social networks with different cultures and educational backgrounds in the last ten years. Various sentiments are expressed in the user comments, from support to hatred. The sentiments regarded the United States General Election in 2020. This dataset has 3,000 data gotten from previous research. We augment it becomes 15,000 data to facilitate training and increase the required data. Sentiment detection is carried out using the CNN-BiLSTM architecture. It is chosen because CNN can filter essential words, and BiLSTM can remember memory in two directions. By utilizing both, the training process becomes maximum. However, this method has disadvantages in the activation. The drawback of the existing activation method, i.e., "Zero-hard Rectifier" and "ReLU Dropout" problem to become the cause of training stopped in the ReLU activation, and the exponential function cannot be set become the activation function still rigid towards output value in the SERLU activation. To overcome this problem, we propose a novel activation method to repair activation in CNN-BiLSTM architecture. It is namely the ASERLU activation function. It can adjust positive value output, negative value output, and exponential value by the setter variables. So, it adapts more conveniently to the output value and becomes a flexible activation function because it can be increased and decreased as needed. It is the first research applied in architecture. Compared with ReLU and SERLU, our proposed method gives higher accuracy based on the experiment results.

Evaluating the Performance of 193 nm Ultraviolet Photodissociation for Tandem Mass Tag Labeled Peptides

Despite the successful application of tandem mass tags (TMT) for peptide quantitation, missing reporter ions in higher energy collisional dissociation (HCD) spectra remains a challenge for consistent quantitation, especially for peptides with labile post-translational modifications. Ultraviolet photodissociation (UVPD) is an alternative ion activation method shown to provide superior coverage for sequencing of peptides and intact proteins. Here, we optimized and evaluated 193 nm UVPD for the characterization of TMT-labeled model peptides, HeLa proteome, and N-glycopeptides from model proteins. UVPD yielded the same TMT reporter ions as HCD, at m/z 126–131. Additionally, UVPD produced a wide range of fragments that yielded more complete characterization of glycopeptides and less frequent missing TMT reporter ion channels, whereas HCD yielded a strong tradeoff between characterization and quantitation of TMT-labeled glycopeptides. However, the lower fragmentation efficiency of UVPD yielded fewer peptide identifications than HCD. Overall, 193 nm UVPD is a valuable tool that provides an alternative to HCD for the quantitation of large and highly modified peptides with labile PTMs. Continued development of instrumentation specific to UVPD will yield greater fragmentation efficiency and fulfil the potential of UVPD to be an all-in-one spectrum ion activation method for broad use in the field of proteomics.

Preparation of Activated Coke by One-step Activation Method, Ammonization and K2CO3 Modification of Coal and Biomass

Coal and biomass are regularly used in the preparation of activated coke. In this paper, coal and biomass (poplar bark) were co-pyrolyzed, and activated coke was prepared by physical “one-step activation”, followed by ammonization and additional activation by potassium carbonate (K2CO3). The activation temperature was set at 800 °C, the time was 60 min, and the activation atmosphere contained 10% steam and 20% CO2 by volume in nitrogen. The physical and chemical properties of activated coke prepared by “one-step activation method” and “two-step activation method” were compared. The mole fraction of ammonia was 5%, 10%, 15%, 20% respectively and the concentration of K2CO3 solution was 3%, 6%, 9%, 12% respectively. It is found that the addition of ammonia can improve physical and chemical properties and adsorption capacity of activated coke, but the increase of ammonia mole fraction has little effect on the activation reaction. The physical and chemical structure of activated coke impregnated with 3% K2CO3 solution was found to be well developed. The surface of the produced activated coke prepared by the above activation methods has abundant microporous structure and organic functional groups, hence the final product is suitable to be used as adsorbent in practical applications.