Against the backdrop of profound structural shifts within the global art-auction ecosystem, small- and medium-sized auction houses (SMEs) are confronted with multi-faceted sourcing dilemmas. These include the growing scarcity of high-quality uncirculated artworks ("fresh lots"), diminishing returns from traditional sourcing channels, mounting compliance burdens brought by updated cultural relic regulations, and persistent competitive disadvantages against internationally leading auction institutions. Conventional sourcing workflows rely heavily on personal social networks and individual expert experience, leading to fragmented information storage and limited capacity to expand beyond existing client circles. Although existing academic papers have explored AI-driven artwork valuation, research focusing on AI-assisted upstream sourcing procedures for resource-constrained SME auction houses remains underdeveloped. This paper adopts the Gioia-style grounded-theory inductive qualitative methodology, drawing on five semi-structured in-depth interviews with core practitioners from auction institutions located in the Yangtze River Delta region of China. Through three-phase coding procedures, four aggregate theoretical dimensions are extracted: systemic root causes of sourcing difficulties, fundamental transformation of channel operation logic, structural differentiation of collector behaviours and mindsets, as well as the value boundaries and practical obstacles for AI implementation. Based on these empirical findings, this study constructs an embedded five-stage closed-loop human-AI collaborative sourcing scheme. Different from high-cost custom-built technical systems, the proposed solution adopts zero-code off-the-shelf tooling, explicitly positioning artificial intelligence as an information-augmentation instrument rather than a complete substitute for human professional judgment and trust-based interpersonal negotiation. This research enriches existing literature concerning digital transformation within the art market and delivers context-aware, actionable operational suggestions for SME auction practitioners with limited technical and financial budgets.

Payment methods now play an increasing role in reshaping mental accounting—especially with regard to ACG merchandise—within a fast-growing part of the cultural economy. A case study of Bilibili Membership Mall is used here to represent Gen Z ACG consumer behavior in China; the analysis relies on both transactional data and survey responses to explore, in some detail, the interaction of digital payments with fandom-related psychology. This study applies mental accounting ideas to the pain-of-paying concept and analyzes specifically, with regard to ACG merchandise, how mobile or credit payment instruments may lead to irrational consumption choices. The work identifies positive roles played by such payment methods: they help to lower consumption barriers, to make consumption more complete, and to promote second-hand use, for these mechanisms also run the risk of creating system-wide problems. With regard to mental accounting distortions or overindebtedness, speculative leveraging on secondary markets, and socially sustained irrational spending. To counter such risks, the present analysis advocates a coordinated governance model—based on platform-level control of payments, secondary-market rules, and localized digital literacy efforts. It examines how modes of payment affect mental accounting and thus expands behavioral finance in ways useful for thinking about the long-term viability of the ACG consumer market.

The rapid expansion of food-delivery platforms has created employment opportunities for millions of workers, yet it has also brought about unstable earnings, inadequate labour protections and intensifying algorithmic control. Existing studies have largely explored workers' earnings from either an individual-level perspective, focusing on training and work experience, or a platform-level perspective, centring on algorithmic management. This review integrates these two perspectives to investigate how learning and algorithmic management shape food-delivery workers' earnings, as well as how workers respond to such working conditions. A literature-review approach was adopted, drawing mainly on three core studies on food-delivery and platform work together with their key cited references. This paper compares scholarship on human capital, learning-by-doing, algorithmic management, worker agency and platform-imposed control. The findings suggest that learning-by-doing and accumulated work experience are generally positively correlated with workers' income, whereas formal platform training may yield limited or even adverse income outcomes when it is decoupled from real-world job demands. Meanwhile, algorithmic management affects earnings via task allocation, performance evaluation, incentive schemes, customer ratings and information asymmetry. Workers acquire practical knowledge and coping strategies to adapt to these systems, yet stronger adaptation does not necessarily translate into greater worker autonomy. This review concludes that food-delivery workers' earnings stem from the interplay between individual learning and platform algorithmic governance. Instead of relying heavily on a one-size-fits-all minimum wage, future interventions ought to prioritise greater algorithmic transparency and accountability. Further research should explore how workers' accumulated experience is recognised and rewarded across different platform systems.

Personalized pricing has become increasingly relevant on digital platforms because firms can use consumer information to estimate differences in willingness to pay and adjust prices accordingly. This review examines how personalized pricing operates, what evidence exists on its effects, and what these findings imply for firms, consumers, and platform governance. The paper first explains the main mechanisms linking consumer information, profiling, demand estimation, and pricing decisions. It then synthesizes evidence from theoretical models, field experiments, consumer experiments, and observations of e-commerce platforms. The reviewed evidence indicates that personalized pricing can improve firm profitability, but its effects on consumers are not uniform. Some consumers may benefit from lower prices, while aggregate consumer surplus may decline. Consumer responses also depend on perceived fairness, privacy concerns, and the degree of price individualization. The review further shows that observed online price differences should be interpreted carefully because differences may arise from factors other than personalized pricing. Overall, personalized pricing should not be understood simply as charging different consumers different prices. Its effects depend on the quality of information, the competitive environment, the degree of personalization, and how consumers perceive the pricing process. The review concludes that future research should place greater emphasis on transaction-level evidence, heterogeneous consumer effects, privacy, and methods that can distinguish genuine personalization from other sources of price variation.

Roulette gives a transparent setting for studying how a small statistical advantage can be converted into a controlled level of repeated financial exposure. This study develops a casino-side analysis based on the Kelly criterion, comparing American and European even-money roulette under full, half, and quarter Kelly strategies. The theoretical analysis derives the growth-maximising exposure from expected logarithmic wealth and uses concavity to establish uniqueness. The model is then calibrated to the two wheel structures and evaluated through Monte Carlo simulation over 1,000 spins. For American roulette, full Kelly produces the highest median terminal bankroll, but also a substantially weaker lower tail and a greater probability of finishing below the initial bankroll than fractional Kelly. European roulette exhibits the same pattern at a smaller scale because its house advantage is lower. These results show that long-run optimality does not remove finite-horizon risk: a strategy can maximise compound growth while still generating deep temporary losses. Fractional Kelly therefore provides a practical compromise when capital constraints, operating expenses, or model uncertainty make downside control important. The roulette benchmark also clarifies how the framework can be extended to blackjack, sports odds, and portfolio allocation, where probabilities and payoffs must be estimated rather than read directly from fixed game rules.

This research investigates the effect of environmental framing on consumers' allocation of expenditure between second-hand clothes and fast fashion goods. To achieve this, a three-stage quantitative online survey was conducted. In the first stage, consumers allocated a sum of RMB 500 across a variety of clothing options. Next, they were either shown materials using a framing approach that stressed the positive influence of second-hand apparel on the environment or a negative frame highlighting the negative effects of fast fashion, and then decided how to allocate the next RMB 500. For evaluation, paired- and independent-samples tests as well as correlation analysis were performed. The outcomes imply that both framing conditions led to more money being allocated to purchasing second-hand clothes. Nevertheless, it was found that there was no difference between the positive and negative frames. Furthermore, moral licensing was not confirmed either: framing that led to a higher allocation for second-hand goods also reduced the amount of money allocated to fast fashion. While total spending increased during this stage of the study, the changes in the budget and choice conditions make it impossible to say that this was a rebound effect.

Under the Mobile Internet, short-video algorithms are now the primary modes of change for culture creation. The two effects of algorithmic recommendations on the logic of content creation and the public's aesthetic demands are examined here. Based on a review of the literature and case studies, and using the algorithmic transformation of Douyin and Kuaishou in 2016-2017 as examples, this paper finds that algorithms employ metrics such as completion rates to create a "traffic pool competition" mechanism. Although it will increase the distribution efficiency in the short run, there will be a loss of content diversity and depth over time. Algorithms limit the scope of users with "cocoons", lower aesthetic standards, and thus reduce interest. Algorithms enhance the segmentation of aesthetics and social groups. Therefore, this paper proposes a four-dimensional optimization system: rearrangement of algorithm parameters, construction of private-traffic brand moats, strengthening platform-based aesthetic guidance, and promotion of public media literacy. This paper will investigate the cultural changes brought about by the algorithmic era and offer some suggestions for platform management, creator development, etc.

Public housing is a major policy instrument addressing housing affordability and socioeconomic inequality in Hong Kong. However, whether the spatial concentration of public housing promotes socioeconomic integration or reinforces the concentration of low-income households remains unclear. This study examines the relationship between public housing concentration and income distribution across the 18 districts of Hong Kong, using district-level data from Hong Kong and an OLS regression. The analysis focuses on whether districts with higher shares of public rental housing (PRH) display systematically different income outcomes. The results show a significant negative relationship between PRH household share and median monthly household income. Districts with higher PRH shares also have significantly higher proportions of low-income households, while robustness checks using alternative measures of income and public housing produce consistent results. Further analysis finds no significant regional heterogeneity between the urban core and New Territories. Overall, the findings suggest a negative association. Future research can combine quantitative analysis with qualitative evidence to better identify the mechanisms underlying this relationship.

The rapid expansion of chain restaurants has become an important growth strategy in China's restaurant industry. However, increasing the number of stores does not necessarily lead to higher-quality growth, as new stores may generate additional demand while also competing with existing stores. This study examines whether changes in store scale affect the growth quality of directly operated chain restaurant brands in China. The study uses the data of selected Hong Kong-listed Chinese restaurant companies from 2021 to 2025 to measure growth quality using same-store sales growth (SSSG) and store expansion using the net store growth rate (NSG). This research use a model to examine whether store expansion in the previous period affects subsequent same-store sales growth, while firm size is included as a control variable. The result supports the cannibalization hypothesis over the brand spillover hypothesis. This study provides empirical evidence on the relationship between expansion strategy and organic growth in China's chain restaurant industry for managers and investors to make decisions.

The COVID-19 pandemic has caused severe disruptions to global supply chains, and the fresh food e-commerce sector has been particularly affected due to the perishable nature of its products and the strict time requirements in distribution. Understanding how fresh food e-commerce platforms cope with such extreme shocks and rebuild their supply chains afterwards is therefore of both practical and research significance. This study takes Hema Fresh as a case to examine the disruption risks it encountered during the pandemic, the emergency measures it adopted, and the strategic transformation it pursued in the post-pandemic period. Using a single-case-study approach and drawing on multiple data sources including corporate financial reports, media coverage, academic papers, and industry research reports, the study applies thematic analysis to identify key patterns in Hema's resilience-building process. The findings show that Hema effectively addressed immediate crises through measures such as the "shared employee" program and multi-source procurement, and later achieved a turnaround from loss to profit through discount reforms and vertical supply chain integration. However, the study also identifies a persistent tension between cost efficiency and resilience redundancy, which remains a key challenge for Hema and the broader fresh e-commerce industry.