Despite the recognized importance of venture capital (VC) in fostering entrepreneurship, its functional evolution remains inadequately conceptualized in existing literature. This paper conducts a systematic review of 109 studies published from 2000 to 2025 and explores the dynamic trajectory of VC in entrepreneurship lifecycle. Moving beyond the traditional view of VC as a mere financial intermediary, we propose a multi-dimensional framework that captures its expanded functions of resource orchestration, strategic partnership, and cognitive collaboration. Our findings indicate that the manifestation of these roles is not uniform, but varies systematically with the type of investor, the timing of their engagement, and the broader institutional setting. By developing a dynamic model of VC functionality, this review provides a nuanced understanding of how VC’s role deepens over time and offers critical insights for entrepreneurs and investors to enhance the efficacy of VC-firm partnerships.

Accurately forecasting hotel booking cancellations is essential for revenue management, operational planning, and customer retention in the hospitality industry. This study provides a comprehensive evaluation of both traditional machine learning models and modern deep learning architectures using the publicly available Hotel Booking Demand dataset. After systematic preprocessing, feature engineering, and handling class imbalance through oversampling and cost-sensitive learning, several algorithms were benchmarked. Among them, ensemble methods such as Random Forest and XGBoost achieved the most reliable results, with overall accuracy of 84% and ROC-AUC scores exceeding 0.91. Deep learning models including CNN, LSTM, and Transformer also demonstrated competitive performance, though they required more computational resources and showed varying sensitivity to data characteristics. Beyond predictive accuracy, SHAP-based interpretability and error analysis highlighted the critical role of features such as lead time, prior cancellations, and number of special requests, offering actionable insights for practitioners. For instance, longer lead times consistently increased cancellation risk, while multiple special requests were strongly associated with lower cancellation probabilities, reflecting guest commitment. The study further emphasizes the importance of minimizing false negatives, as misclassified cancellations lead directly to lost revenue. Business-oriented strategies such as dynamic pricing, targeted loyalty programs, and seasonal model adjustment are proposed to reduce risks and improve operational outcomes. Overall, this research confirms the value of cost-aware and interpretable machine learning approaches in optimizing hotel booking management, while also outlining future directions for integrating loyalty profiles and customer feedback into predictive systems.

With the wide application of artificial intelligence and big data technologies, personalized advertising has become a mainstream strategy for digital marketing. However, whether highly relevant ad recommendations always lead to better user acceptance is increasingly being questioned. This study investigates the impact of ad recommendation relevance (strong vs. weak) on user acceptance and examines the moderating role of users' exploratory tendency. Using a simulated ad experiment combined with an online survey, it finds that while strongly relevant ads generally receive higher acceptance, weakly relevant ads are more attractive to users with higher exploratory tendencies. Privacy concerns and interest domain heterogeneity also influence ad effectiveness. This research contributes to optimizing ad resource allocation and enhancing the return on investment in advertising.

Against the dual dropout of the continuous recovery of the global tourism industry and the goal of carbon neutrality, how online travel platforms balance economic benefits and environmental responsibilities has become a key issue in the industry. Focusing on the business performance of Trip.com Group in 2024, this paper conducts an in-depth analysis from the dual dimensions of financial data and green environmental protection practices. Financially, in 2024, Trip.com Group's net revenue was approximately 53.377 billion yuan, a year-on-year increase of about 19.78%, and its net profit attributable to the parent company was approximately 17.067 billion yuan, a year-on-year increase of about 72.08%. All core business segments experienced varying degrees of growth. In terms of green environmental protect ion, Trip.com actively implemented a sustainable development strategy, introduced the "Low-Carbon Hotel Standard", and drove over 16 million users to choose low-carbon travel. Through a study of the correlation between the two, this paper reveals Trip.com's exploration of an environmentally friendly development model while pursuing economic growth, providing a reference for the development of the online travel industry.

The digital economy is expanding rapidly, profoundly reshaping global labour markets and presenting both opportunities and challenges for female labour force participation. Although existing literature has explored this relationship, the academic landscape remains fragmented. This study offers a systematic overview through bibliometric analysis, drawing on a dataset of 552 publications retrieved from the Web of Science (WOS) (1950-2025). Using VOSviewer, this study conducts co-authorship, citation, and co-occurrence analyses. The results reveal a surge in publications since 2017, with the USA, China, and England forming the core collaborative network. Citation analysis identifies foundational literature and key journals, while keyword co-occurrence clustering uncovers four dominant research themes: the gendered labour-time dilemma, the autonomy paradox in the gig economy, trust differences in e-commerce, and gender-differentiated innovation during crises. This study seeks to consolidate the knowledge architecture and provide a foundational reference for future scholarship on gender and digital transformation.

Under the background of increasing volatility in the financial market, the cross day jump poses a challenge to the threshold framework of the traditional paired trading strategy. This study distinguishes between intraday fluctuations and inter day jumps, uses Bayesian search to dynamically adjust the threshold, and combines sliding window and co-integration theory to build a medium and high frequency quantitative strategy, filling the gap in the research of medium and high frequency paired trading in emerging trading markets such as Kechuang 50. The empirical results in the 50 sector of science and technology innovation show that the strategy has both high-yield and risk control capabilities. After deducting the handling fees, the performance of the strategy is better than that of index investment, and the sharp rate has increased by 107%. However, the strategy is restricted by short selling of a shares, requires securities lending and has a high threshold, which is only applicable to large institutions; At the same time, there is a waste of handling fees caused by the overlapping of stock pairs. In the future, we can improve the unilateral strategy to adapt to market constraints, and introduce external data such as big language model and financial news to alleviate the lag of strategy adjustment.

This paper examines the impact of the UK's R&D tax relief reform on its biotech and pharmaceutical industries. As a global R&D leader with pharmaceutical investment exceeding 25% of total R&D expenditure, the UK's 2023 policy restricting relief to domestic activities creates both opportunities and challenges. The analysis demonstrates that tax relief significantly reduces R&D costs, enhances corporate financial indicators, and stimulates innovation investment. However, limitations on outsourcing increase operational costs and restrict international collaboration. While the policy strengthens local R&D ecosystems and attracts investment, it simultaneously hinders global knowledge exchange. The study concludes that despite implementation challenges, the reform's benefits outweigh its drawbacks. Recommendations include industrial adaptation through R&D expansion and governmental policy optimization to balance domestic innovation with global cooperation.

This study develops an analytical tool to price the inclusive finance considering the marginal risk contributions of the borrower within the prudential bounds defined by regulatory bodies. The empirical study uses data from 60 institutions involved in the process of inclusive finance in Asia and Africa between 2015 and 2024, to perform stress tests using fixed effect regression models to examine the performance in compliance and portfolio risks. The findings demonstrate that the risk-contribution pricing model has resulted in decreasing portfolios' volatility by 18.7 ± 1.4% in contrast to the conventional models, with default clustering decreasing by 20.2 ± 1.1% and inclusiveness increasing by 9.3 ± 0.8%. The regression model validates the significance of the marginal risk contributions at the borrower level in understanding and impacting the default probability (β = 0.417, t = 6.91, p < 0.001) and inclusiveness ratios (β = 0.281, t = 4.18, p = 0.003). The findings of the study validate that with the application of the proposed model in accordance with the tightening of Basel III rules, inclusiveness can be maintained along with sustainability in the context of prudential regulation in the field of inclusive finance.