Your search

Stock price prediction has always been challenging due to its volatility and unpredictability. This paper performs a preliminary exploratory comparison that utilizes Long Short-Term Memory (LSTM) and Support Vector Machine (SVM) algorithms to forecast the stock market in Hong Kong. It considers a public dataset publicly available and uses feature engineering to extract relevant features. Then, LSTM and SVM algorithms are applied to predict stock prices. Our results show that the proposed machine learning techniques can predict stock prices in Hong Kong's share market with the error metrics presented, and, for this purpose, LSTM achieved better results than SVM, with MSE = 0.0026, RMSE = 0.0508, MAE = 0.0406, and MAPE = 1.325.

This dissertation identifies factors driving consumer shopping behavior within the realm of live-streaming commerce, an area fast emerging in the e-commerce domain. Live-streaming shopping or commerce involves real-time interaction and entertainment with traditional online shopping, forming a unique endogenous environment where consumers can contact sellers or influencers directly. The study employed quantitative surveys that identified some of the main determining factors of consumer behavior within this context. The findings show that the significant factors in determining consumer behavior are trust and engagement, which are strongly influenced by the credibility and authenticity of the live streamer. Another significant finding is the role of social interaction and community building in providing consumers with a sense of belonging and validation, enhancing their confidence and purchase intention. Moreover, it highlights how marketing strategies of flash sales, limited-time offers, and partnerships with influencers make their way into the system to help invoke engagement and impulsive buying behavior among consumers. The implications of these findings extend to e-commerce platforms and marketers. Any improvements in features leading to trust, engagement, and interactivity within the community would drive higher customer satisfaction and sales. According to researchers, working partnerships with believable influencers and more extensive integrations of real-time marketing might further activate live-streaming commerce. This study thus fills a gap in the existing body of literature by detailing the drivers of consumer behavior toward live-stream commerce. It also identifies areas of future research on the current studies, including developing technologies and the cultural variances in the impact of live-stream commerce, including ethical considerations. These results are principle for guiding work on potential live-stream commerce in the digital age for anybody from workers to academicians

The manifestation of generating digital visuals through an algorithm is gaining worldwide attention in the graphic design industry. It is a new form of computing that visualizes data input by the designer or collected in the physical environment and turns them into artwork. The generative design of...

Macau has long been considered to be an example of remarkable economic growth. With the opening of the gaming sector in 2002, the casino and hospitality sector flourished, creating employment opportunities but also imposing several challenges on managers. Since Macau endeavors to be positioned as the center for international business with Portuguese-speaking countries and a platform for trading with China’s Greater Bay Area (GBA), it becomes essential for international enterprises to understand the local dynamics. In light of the limited research available, this study aims to identify management challenges from the perspectives of senior executives in different industries based in Macau. Our findings point out that managers must contend with several issues, such as the lack of a skilled local talent pool, high turnover rates, employees' work attitudes, and a tightly controlled immigration policy. It is also imperative for international managers to nurture relationships and pay attention to the local culture. Our results suggest that Macau has to develop a highly skilled local workforce to attract international companies, while local organizations also have to create an attractive working environment to compete in the marketplace.

A combination of assessment, operational forecast, and future perspective was thoroughly explored to provide an overview of the existing air quality problems in Macao. The levels of air pollution in Macao often exceed those recommended by the World Health Organization (WHO). In order for the population to take precautionary measures and avoid further health risks during high pollution episodes, it is important to develop a reliable air quality forecast. Statistical models based on linear multiple regression (MLR) and classification and regression trees (CART) analysis were successfully developed for Macao, to predict the next day concentrations of NO2, PM10, PM2.5, and O3. Meteorological variables were selected from an extensive list of possible variables, including geopotential height, relative humidity, atmospheric stability, and air temperature at different vertical levels. Air quality variables translate the resilience of the recent past concentrations of each pollutant and usually are maximum and/or the average of latest 24-hour levels. The models were applied in forecasting the next day average daily concentrations for NO2 and PM and maximum hourly O3 levels for five air quality monitoring stations. The results are expected to support an operational air iv quality forecast for Macao. The work involved two phases. On a first phase, the models utilized meteorological and air quality variables based on five years of historical data, from 2013 to 2017. Data from 2013 to 2016 were used to develop the statistical models and data from 2017 was used for validation purposes. All the developed models were statistically significantly valid with a 95% confidence level with high coefficients of determination (from 0.78 to 0.93) for all pollutants. On a second phase, these models were used with 2019 validation data, while a new set of models based on a more extended historical data series, from 2013 to 2018, were also validated with 2019 data. There were no significant differences in the coefficients of determination (R2) and minor improvements in root mean square errors (RMSE), mean absolute errors (MAE) and biases (BIAS) between the 2013 to 2016 and the 2013 to 2018 data models. In addition, for one air quality monitoring station (Taipa Ambient), the 2013 to 2018 model was applied for two days ahead (D2) forecast and the coefficient of determination (R2) was considerably less accurate to the one day ahead (D1) forecast, but still able to provide a reliable air quality forecast for Macao. To understand if the prediction model was robust to extreme variations in v pollutants concentration, a test was performed under the circumstances of a high pollution episode for PM2.5 and O3 during 2019, and a low pollution episode during 2020. Regarding the high pollution episode, the period of the Chinese National Holiday of 2019 was selected, in which high concentration levels were identified for PM2.5 and O3, with peaks of daily concentration for PM2.5 levels exceeding 55 μg/m3 and the maximum hourly concentration for O3 levels exceeding 400 μg/m3. For the low pollution episode, the 2020 period of implementation of the preventive measures for COVID-19 pandemic was selected, with a low record of daily concentration for PM2.5 levels at 2 μg/m3 and maximum hourly concentration for O3 levels at 50 μg/m3. The 2013 to 2018 model successfully predicted the high pollution episode with high coefficients of determination (0.92 for PM2.5 and 0.82 for O3). Likewise, the low pollution episode was also correctly predicted with high coefficients of determination (0.86 and 0.84 for PM2.5 and O3, respectively). Overall, the results demonstrate that the statistical forecast model is robust and able to correctly reproduce extreme air pollution events of both high and low concentration levels. Machine learning methods maybe adopted to provide significant improvements in combination of multiple linear regression (MLR) and classification and regression vi tree (CART) to further improve the accuracy of the statistical forecast. The developed air pollution forecasting model may be combined with other measures to mitigate the impact of air pollution in Macao. These may include the establishment of low emission zones (LEZ), as enforced in some European cities, license plate restrictions and lottery policy, as used in some Asian, tax exemptions on electric vehicles (EVs) and exclusive corridors for public transportations. Keywords: Air pollution; Particulate Matter; Ozone; Macao; Statistical air quality forecast; Pollution episodes; Chinese national holiday; COVID-19

The rapid development of counselling began after World War II. Meanwhile, texting services became available in the 1980s. Texting has been widely applied in both health area and mental health areas. A chatbot is a specific type of artificial intelligence (AI) that can have conversations with humans through texting. The combination of counselling and texting is known as chat counselling. The effectiveness of chatbots in alleviating psychological symptoms was supported by scientific research. Nonetheless, people hold different perceptions towards chatbots. Some common factors in counselling, including empathy and warmth, were also important in evaluating AI chatbots. Other important factors included acceptability, satisfaction and trust. The current study is the first to use AI as a counsellor. This study investigates people’s perceptions of human and AI counsellors (ChatGPT) and whether people can differentiate between human and AI. Participants needed to rate the counsellors in three scenarios: the original scenario taken from a training book, the human counsellor scenario generated from a text conversation with a human counsellor and the AI counsellor scenario produced by texting with ChatGPT, which acts as the counsellor. Prompts used to generate conversation with ChatGPT are included. The dialogues were parts of the conversations containing similar client responses and were presented using the WhatsApp interface. Questionnaires were delivered both online and in paper form. Results demonstrated that people’s ratings of human counsellors and AI counsellors did not differ in perceived empathy, acceptability, and satisfaction. While the warmth and trustworthiness of AI counsellors were perceived to be higher than those of human counsellors. On the other hand, people were unable to differentiate between human and AI counsellors in uncertain conditions. Younger people and the general population are more accurate in identifying between humans and AI, while people above 40 and psychologists or counsellors are less capable of doing so. The current study supports the potential of utilising ChatGPT in counselling. Having people experience and evaluate real chat counselling with human and AI counsellors can potentially eliminate some limitations of the current studies. Future studies can investigate how prior knowledge contributes to AI detection and examine AI counsellors' efficiency in longitudinal studies

The area of clinical decision support systems (CDSS) is facing a boost in research and development with the increasing amount of data in clinical analysis together with new tools to support patient care. This creates a vibrant and challenging environment for the medical and technical staff. This chapter presents a discussion about the challenges and trends of CDSS considering big data and patient-centered constraints. Two case studies are presented in detail. The first presents the development of a big data and AI classification system for maternal and fetal ambulatory monitoring, composed by different solutions such as the implementation of an Internet of Things sensors and devices network, a fuzzy inference system for emergency alarms, a feature extraction model based on signal processing of the fetal and maternal data, and finally a deep learning classifier with six convolutional layers achieving an F1-score of 0.89 for the case of both maternal and fetal as harmful. The system was designed to support maternal–fetal ambulatory premises in developing countries, where the demand is extremely high and the number of medical specialists is very low. The second case study considered two artificial intelligence approaches to providing efficient prediction of infections for clinical decision support during the COVID-19 pandemic in Brazil. First, LSTM recurrent neural networks were considered with the model achieving R2=0.93 and MAE=40,604.4 in average, while the best, R2=0.9939, was achieved for the time series 3. Second, an open-source framework called H2O AutoML was considered with the “stacked ensemble” approach and presented the best performance followed by XGBoost. Brazil has been one of the most challenging environments during the pandemic and where efficient predictions may be the difference in saving lives. The presentation of such different approaches (ambulatory monitoring and epidemiology data) is important to illustrate the large spectrum of AI tools to support clinical decision-making.