Aspect-level sentiment classification is a fine-grained task in sentiment analysis. In recent years, researchers have realized the importance of the relationship between aspect term and sentence and many classification models based on deep learning network have been proposed. However, these end-to-end deep neural network models lack flexibility and do not consider the sentiment word information in existing methods. Therefore, we propose a lexicon-enhanced attention network (LEAN) based on bidirectional LSTM. LEAN not only can catch the sentiment words in a sentence but also concentrate on specific aspect information in a sentence. Moreover, leveraging lexicon information will enhance the model's flexibility and robustness. We experiment on the SemEval 2014 dataset and results find that our model achieves state-ofthe-art performance on aspect-level sentiment classification. INDEX TERMS Natural language processing, sentiment analysis, aspect-level, sentiment lexicon, attention mechanism.
Abstract:In traditional information technology project portfolio management (ITPPM), managers often pay more attention to the optimization of portfolio selection in the initial stage. In fact, during the portfolio implementation process, there are still issues to be optimized. Organizing cooperation will enhance the efficiency, although it brings more immediate risk due to the complex variety of links between projects. In order to balance the efficiency and risk, an optimization method is presented based on the complex network theory and entropy, which will assist portfolio managers in recognizing the structure of the portfolio and determine the cooperation range. Firstly, a complex network model for an IT project portfolio is constructed, in which the project is simulated as an artificial life agent. At the same time, the portfolio is viewed as a small scale of society. Following this, social network analysis is used to detect and divide communities in order to estimate the roles of projects between different portfolios. Based on these, the efficiency and the risk are measured using entropy and are balanced through searching for adequate hierarchy community divisions. Thus, the activities of cooperation in organizations, risk management, and so on-which are usually viewed as an important art-can be discussed and conducted based on quantity calculations.
Nano-TiO2 has always been one of the most important topics in the research of photocatalysts due to its special activity and stability. However, it has always been difficult to obtain nano-TiO2 with high dispersion, a small particle size and high photocatalytic activity. In this paper, nano-TiO2 powder was prepared by combining the high-gravity technique and direct precipitation method in an impinging stream-rotating packed bed (IS-RPB) reactor followed by Fe3+ in-situ doping. TiOSO4 and NH3·H2O solutions were cut into very small liquid microelements by high-speed rotating packing, and the mass transfer and microscopic mixing of the nucleation and growth processes of nano-TiO2 were strengthened in IS-RPB, which was beneficial to the continuous production of high quality nano-TiO2. Pure TiO2 and iron-doped nano-TiO2 (Fe-TiO2) were obtained in IS-RPB and were investigated by means of X-ray diffraction (XRD), Raman, scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), ultraviolet-visible diffuse reflectance spectroscopy (UV-vis DRS) and Brunauer–Emmett–Teller (BET) analysis, which found that pure TiO2 had a particle size of about 12.5 nm, good dispersibility and a complete anatase crystal at the rotating speed of packing of 800 rpm and calcination temperature of 500 °C. The addition of Fe3+ did not change the crystalline structure of TiO2. Iron was highly dispersed in TiO2 without the detection of aggregates and was found to exist in a positive trivalent form by XPS. With the increase of iron doping, the photoresponse range of TiO2 to visible light was broadened from 3.06 eV to 2.26 eV. The degradation efficiency of gaseous toluene by Fe-TiO2 under ultraviolet light was higher than that of pure TiO2 and commercial P25 due to Fe3+ effectively suppressing the recombination of TiO2 electrons and holes; the highest efficiency produced by 1.0% Fe-TiO2 was 95.7%.
At present, robot embedded systems have some common problems such as closure and poor dynamic evolution. Aiming at resolving this situation, our paper focuses on improvements to the robot embedded system and sets up a new robot system architecture, and we propose a syncretic mechanism of a robot and SoftMan (SM). In the syncretic system, the structural organization of the SoftMan group and its modes are particularly important in establishing the task coordination mechanism. This paper, therefore, proposes a coordination organization model based on the SoftMan group, and studies in detail the process of task allocation for resource contention, which facilitates a rational allocation of system resources. During our research, we introduced Resource Requirement Length Algorithm (RRLA) to calculate the resource requirements of the task and a resource conformity degree allocation algorithm of Resource Conformity Degree Algorithm (RCDA) for resource contention. Finally, a comparative evaluation of RCDA with five other frequently used task allocation algorithms shows that RCDA has higher success and accuracy rates with good stability and reliability.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.