Search

Many studies have been proposed to prove that technical analysis can help investors make trading decisions. The moving average (MA) is a widely used technical indicator that plays an important role in this field since it directly reflects stock fluctuations. However, most studies ignore the parameter settings of the MA, which leads to underestimation of the potential of the MA. Therefore, this paper is the first attempt to remove all restrictions and extend the limits of the MA to take advantage of the MA’s capability well. It also uses different kinds of MA, such as the weighted moving average (WMA) and the exponential moving average (EMA), to compose trading strategies. Our system proposes the global best-guided quantum-inspired tabu search algorithm (GQTS), which is better at searching than traditional algorithms, to optimize trading strategies based on the MA. Furthermore, an innovative 2-phase sliding window is invented to consider more investment situations in changeable stock markets. In summary, this paper intended to investigate the ability of MA and proposed dynamic and intelligent trading strategies based on MA, GQTS, and 2-phase sliding window to assist investors to make trading decisions. The experiments show that the proposed system flexibly discovers better trading points. Our system outperforms traditional methods and beats the buy-and-hold strategy to yield significant profits both in developed and emerging stock markets.

Nowadays, there are numerous social platforms for people to discuss and share information. However, certain users may post contents that discriminate against specific groups or spread inappropriate behaviors. Therefore, it is desired to have a detection mechanism that is able to accurately identify these offensive comments and inappropriate messages before appearing on social platforms. To address this need, we propose a novel model called Attention-based Multi-filter Convolutional Neural Network (AMCNN) for inappropriate speech detection. Our model combines CNN and BiLSTM to acquire the advantages of the two. AM CNN leverages CNN filters of various kernel sizes for convolution operations to capture the relation between words of different lengths. The convolutional features are then used as the inputs of BiLSTM, which is able to obtain the bidirectional semantics via the forward and backward directions. AMCNN further incorporates the attention mechanism to highlight the important parts of the hidden layer output for better classification. Our model can be applied to various types and lengths of text, and it is also able to support different languages. To evaluate the model performance, we conduct extensive experiments on various real-world datasets. According to the experimental results, our model achieves the best performance, which is on average 12.63% higher than the state-of-the-art approach on the F1-score indicator. We also conduct ablation analysis, and the results show that the attention mechanism and the usage of BiLSTM can considerably improve the classification performance and effectively detect inappropriate speech in sentences.

Evolutionary algorithms or metaheuristic methods are common approaches applied in highly complex optimization problems, such as stock trading. In this paper we employ a novel approach for using a dynamic trading system with a technical indicator called the moving average (MA). Moreover, we utilize an improved metaheuristic algorithm, the global-best guided quantum-inspired tabu search algorithm (GQTS), which has a fast and stable feature to efficiently search for the optimal trading strategy of MA. Our approach employs the sliding window technique to avoid overfitting problem. In addition, we propose year-on-year sliding windows and 2-phase sliding window to adapt to the phenomenon of an economic cycle in the investment period and to efficiently solve stock trading problems. The experimental environment is the United States stock market. The experiment result shows that combine GQTS and MA can find better strategies that outperform other normal strategies. The performance of remuneration indicates that the trading system is enhanced with a year-on-year sliding window and 2-phase sliding window.

Selecting a portfolio with high return and low risk is a difficult problem that is worthy of research. When solving a portfolio optimization problem, research studies usually employ various assessing indicators. This paper utilizes the trend ratio as the measure indicator. The value of the trend ratio represents how much the return is from the portfolio per unit of risk. A portfolio selected by the trend ratio based on simple linear regression is more in accordance with investors’ psychology. This paper proposes a new investment strategy with fund allocation to solve a portfolio optimization problem, as fund allocation can make the portfolio more flexible and also can reduce risk effectively. This paper uses the Global-best Guided Quantum-inspired Tabu Search with Quantum-NOT Gate (GNQTS) to find how many funds are allocated to each stock and then uses the trend ratio to evaluate the portfolio. Because overfitting is a common problem in the stock market, this paper uses thirteen types of sliding windows to avoid overfitting. The results show that fund allocation is more flexible than allocating equal funds, thus allowing the proposed method to find a portfolio with a higher return and lower risk.

This paper study, we present a design solution for rectangular waveguide filters suitable for high-power applications. To verify the proposed filter, the S-parameter characteristics of the waveguide filter were compared with other types of filters by electromagnetic simulation. The simulation data shows that the proposed filter solution can be used in the limited space of high-power radar system applications.

Selecting the best combination of stocks with low risk and high return simultaneously is a significant challenge for investors. The Sharpe ratio is a common way to evaluate the performance of portfolios. However, the Sharp ratio has some defects such as that portfolios in uptrend will be calculated as high risk. Therefore, trend ratio on linear regression is proposed to find the portfolio that has a stable uptrend. Since the linear regression just roughly represents the portfolio trend and cannot be approached precisely, this paper proposes the quadratic regression trend to define the portfolio trend more accurately. In addition, the trend ratio on quadratic regression can signify the change of magnitude slowing down or speeding up in an uptrend or downtrend direction to help this model predict the future trend more precisely. This paper uses the adaptive quantum-inspired tabu search with not gate (ANQTS) to search the best portfolio and employs the 2-phase sliding window to consider more investment situations. The experiment results show that the proposed portfolio optimization model has better performance than the Sharp ratio and trend ratio on linear regression.

Integration of omni-directional transponder antennas into unmanned aerial vehicles (UAVs) becomes an important task since the adjacent conducting elements and UAV payloads significantly degrade the antenna performance. In this paper, an electromagnetic compatible (EMC) transponder antenna for tactical automatic landing system (TALS) is presented. The proposed transponder antenna has a monopole-like radiator designed to operate in X-band for UAV guidance applications, and is backed by a shielding metal cover to suppress the unwanted radio-frequency interference. In addition, with the use of the streamlined dielectric cover, the aerodynamic drag of the antenna can be mitigated. To substantiate the proposed antenna design, an electromagnetic simulation software that uses moment method (MoM) numerical computation and a 3D simulation model of large-sized military UAV are employed. Electromagnetic coupling mechanisms between the antenna and the adjacent conducting devices are also studied for the development of layout rules for transponder antenna.

Metaheuristic algorithms are implemented to solve optimization problems and have recently received significant research attention. Metaheuristic algorithms rely primarily on two properties, exploration, and exploitation. Traditional metaheuristic algorithms use many weights (parameters) to balance these two properties to increase the chance of finding a better solution in limited cost and time. However, traditional algorithms have some problems. Exploration and exploitation are different abilities and restrict each other, therefore, traditional algorithms need many parameters and lots of costs to achieve the balance, and also need to adjust parameters for different optimization problems. Jaguar Algorithm (JA) has great abilities both in exploitation and exploration, is proposed to address these issues. First, JA attempts to find the optimal solution in the designated search area. It then uses history information to jump to a better area. JA can, therefore, determine the position of the global optimum. JA achieves strong exploitation and exploration with these features. Also, according to different problems, JA implements adaptive parameter adjustment. The self-analysis and experiment of this research demonstrate that each JA capability can have various positive effects, while the performance comparison demonstrates JAs superiority over traditional metaheuristic algorithms.