Toward Fair Recommendation in Two-sided Platforms

We investigate the problem of fair recommendation in the context of two-sided online platforms, comprising customers on one side and producers on the other. Traditionally, recommendation services in these platforms have focused on maximizing customer satisfaction by tailoring the results according to the personalized preferences of individual customers. However, our investigation reveals that such customer-centric design may lead to unfair distribution of exposure among the producers, which may adversely impact their well-being. On the other hand, a producer-centric design might become unfair to the customers. Thus, we consider fairness issues that span both customers and producers. Our approach involves a novel mapping of the fair recommendation problem to a constrained version of the problem of fairly allocating indivisible goods. Our proposed FairRec algorithm guarantees at least Maximin Share (MMS) of exposure for most of the producers and Envy-Free up to One item (EF1) fairness for every customer. Extensive evaluations over multiple real-world datasets show the effectiveness of FairRec in ensuring two-sided fairness while incurring a marginal loss in the overall recommendation quality.

Major online platforms today can be thought of as two-sided markets with producers and customers of goods and services. There have been concerns that over-emphasis on customer satisfaction by the platforms may affect the well-being of the producers. To counter such issues, few recent works have attempted to incorporate fairness for the producers. However, these studies have overlooked an important issue in such platforms -- to supposedly improve customer utility, the underlying algorithms are frequently updated, causing abrupt changes in the exposure of producers. In this work, we focus on the fairness issues arising out of such frequent updates, and argue for incremental updates of the platform algorithms so that the producers have enough time to adjust (both logistically and mentally) to the change. However, naive incremental updates may become unfair to the customers. Thus focusing on recommendations deployed on two-sided platforms, we formulate an ILP based online optimization to deploy changes incrementally in η steps, where we can ensure smooth transition of the exposure of items while guaranteeing a minimum utility for every customer. Evaluations over multiple real world datasets show that our proposed mechanism for platform updates can be efficient and fair to both the producers and the customers in two-sided platforms.

Nonmedical Costs of Colorectal Cancer Screening With the Fecal Occult Blood Test and Colonoscopy

Personalised service recommendation is the key technology for service platforms; the demand preferences of users are the important factors for personalised recommendation. First, in order to improve accuracy and adaptability of service recommendation, services are needed to be initialised before being recommended and selected, then they are classified and clustered according to demand preferences, and service clusters are defined and demonstrated. In the sparse problem of service function matrix, historical and potential preferences are expressed as double matrices. Second, service cluster is viewed as the basic business unit, we optimise graph summarisation algorithm and construct service recommendation algorithm SCRP, helped by the experiments about variety parameters, which has more advantages than other algorithms. Third, we select fuzzy degree and difference to be the two key indicators, and use some service clusters to complete simulating and analyse algorithm performances. The results show that our service selection and recommendation method is better than others, which might effectively improve the quality of service recommendation.

Customers in today's marketplace are more demanding, not just of products, but also of service. For manufacturing industries, having a good understanding of the customer's preferences is a key issue to improve customer's satisfaction. Although many approaches for gaining customer's product preferences are reported in the literatures, the problem of analysis customer's service preferences is still not well addressed. To that end, a fuzzy analytic network process (F-ANP) model for measuring the customer's service preferences is proposed in this paper. Firstly, the service factors are summarized under three sales stages. Secondly, based on the complex relationships between the service factors and the customers, the decision network is constructed, and the F-ANP decision process is developed. Finally, a case study is provided to demonstrate the effectiveness of the proposed model.

By introducing the optimal linear predictive code technique into the dynamic issue of lossless image compression, this paper presented a less complexity fitness function for genetic programming engine, which can reduce the cost of computational time in each evaluation for individual greatly, and can also provide further benefit with the scalability issue. To make the speed of large image compression faster in condition of not increasing the cost of computational resource and time, evaluating mechanism in the field of machine learning was used to help genetic programming, and the scalability issue was mapped to the task of making the approach accuracy best from lower speed sampling to higher speed sampling in the field of signal processing. In experiments for compressing large images, the cost of computational time was reduced evidently and efficiently.

The user-centered service paradigm has attracted extensive attention from academia and industry. It advocates taking customer requirements as the orientation, maximizing customer satisfaction as the objective, then targeting to carry out market segmentation and multi-level SLA customization. There are two main challenges: personalized preferences of large-scale customers and resource constraints. In this paper, we propose a resource-constrained multi-level SLA customization approach based on QoE analysis of large-scale customers. With a deep generative network, we fit satisfaction mapping functions and infer the customer’s personalized preference interval for each QoS. Then, based on the theory of granular computing, a multi-level, multi-perspective and multi-scale granular structure for service customization is constructed. Finally, the best match between users with personalized preferences and resources with differentiated qualities is mined to obtain a reduced and balanced multi-level SLA customization scheme. This paper conducts experiments based on the real data of a hotel booking platform and proves that the method performs well in service customization granularity, preference coverage and matching accuracy. The method is an on-demand optimization and avoids over-optimization. The final customized solutions can not only meet the personalized preferences but also give play to the advantages of different quality resources.

Customer service plays a crucial role in shaping customer satisfaction, particularly in the competitive café industry. This study explores the relationship between customer service quality and customer satisfaction in selected cafés in Cagayan de Oro City. This aims to look at the various aspects of service delivery, customer preferences, and the factors that contribute to overall satisfaction levels. The quality of interactions, the consideration of staff, and the overall ambiance are key factors in customer happiness and repeat encouragement. This study utilized random sampling and were based on the total average population of 716 regular customers from the 5 selected cafes in Uptown North wing in Cagayan de Oro City and the list will be taken from the different managers in the selected cafes. In this manner, the researchers were able to determine the sample size of 257 respondents using the Sloven formula by Sloven (1960) the total sample size of 257 respondents wherein, Café A 65 customer respondents, Café B 88 customer respondents, Café C 31 customer respondents, Café D 32 customer respondents. Using a descriptive research design, data were analyzed through descriptive and inferential statistics using mean and standard deviation, inferential statistics, T-test, and ANOVA in determining the significant difference, Person R was used to test the significant relationship. Findings revealed that the overall mean of customer service is 3.49 (3.47+3.50+3.49) with interpretation of “Very good customer service”. The overall mean of customer satisfaction is 3.55 (3.53+3.55+3.58) with an interpretation of “Very high customer satisfaction”. This correlation is categorized as "Strong," and the p-value indicates a "Highly Significant" association, per the legend provided. Nonetheless, a correlation value of 0.708745 indicates that customers happiness tends to rise in tandem with improvements in customer service quality. The study focuses on how customer service affects consumer satisfaction. According to the statistics, customer satisfaction in terms of customer expectations, perceived value, and service quality is significantly correlated with customer service in terms of responsiveness, reliability, and empathy. This research highlights the growing expectations of customers in today’s competitive café industry, where quality service is as essential as a food and beverages industries.

Purpose This project report delves into the customer preferences towards banking services provided by Union Bank of India in Guntur. Recognizing the competitive landscape of the banking sector, this study aims to uncover the key factors that influence customer satisfaction and preferences, thereby offering insights for enhancing service delivery. To achieve this, we surveyed 150 customers of Union Bank of India in Guntur, gathering data on various aspects such as demographics, infrastructure, security, technology, and customer service. The analysis was performed using SPSS software, incorporating a range of statistical tests including Cronbach's alpha for reliability, multiple regression analysis, descriptive statistics, and correlation tests. Design/Methodology/Approach: This research adopts a mixed-methods approach, combining quantitative and qualitative data. Quantitative data will be collected via surveys and financial performance analyses of blue economy organizations, while qualitative data will be obtained through interviews with key stakeholders, such as investors, policymakers, and industry experts. This dual approach aims to provide a holistic understanding of sustainable investment behaviors. Findings: The findings reveal that technology, infrastructure, security, and customer service significantly impact customer preferences. Specifically, technological integration in banking services emerged as the most influential factor, highlighting the growing importance of digital solutions in customer satisfaction. Originality/Value: Additionally, infrastructure and security were found to play critical roles in shaping customer preferences, underscoring the need for robust and secure banking environments. Customer service quality also demonstrated a notable influence, affirming the importance of personalized and efficient service in retaining and attracting customers Research Limitations/Implications: This report not only provides a comprehensive analysis of the factors affecting customer preferences but also offers actionable recommendations for Union Bank of India to enhance their service offerings. By prioritizing technological advancements, improving infrastructure, ensuring security, and maintaining high standards of customer service, the bank can better meet the evolving needs of its customers in Guntur. Practical Implications: The insights derived from this study are invaluable for banking professionals, policymakers, and researchers aiming to understand and improve customer satisfaction in the banking sector. This report contributes to the existing body of knowledge and serves as a practical guide for optimizing banking services to align with customer expectations. Social Implications: The evolution of banking services from traditional banking to the digital era has been marked by several key milestones that have fundamentally transformed the industry. Traditionally, banking services were primarily conducted through physical branches, where customers engaged in face-to-face interactions for tasks such as deposits, withdrawals, and loan applications. The introduction of ATMs in the 1960s revolutionized access to cash, providing customers with convenience and flexibility. Keywords: Digital Banking, Mobile Banking, Online Transactions, Fintech Integration, Customer Experience (CX), Personalization

MapReduce is an important programming model for processing big data with a parallel, distributed algorithm on a cluster. In big data analytic application, equi-join is an important operation. However, it is inefficient to perform equi-join operations in MapReduce when multiple datasets are involved in the join. In this paper, a time cost evaluation model is extended for an equi-join by considering the time cost of calculation. In addition, the sub-joins in an equi-join are classified into star pattern sub-joins on single attribute and chain pattern sub-joins. Based on the extended model, optimization methods are presented and an equi-join plan with lower time cost is chosen for the equi-join. The optimization methods include: the star pattern sub-joins on one attribute are first processed; next, a chain pattern sub-join with minimal scale of intermediate results (i.e. the number of tuples in intermediate results) is processed; at last, a chain pattern sub-join is decomposed into several MapReduce jobs or single MapReduce job by dynamic programming to obtain an optimal scheme for the chain pattern sub-join. We conducted extensive experiments, and the results show that our method is more efficient than those methods such as MDMJ, Hive and Pig.

The location problem of unmanned vending machine is challenging due to the variety of customer preferences and random breakdown in service. In this paper, we present an optimization model for reliable location design of unmanned vending machines, with the goal to minimize total costs and maximize customer satisfaction. We solve the problem through a two-stage approach in order to mine customers preference from their behaviours and improve design reliability. At the first stage, we design a multi-dimensional measurement to mine customers’ preferences and satisfaction based on their behavioural information. At the second stage, we use a clustering method to analyse the set of candidate points from a systematic perspective. Candidate points with similar locations and customer preferences will be clustered into one “demand zone” and the mutual rescue strategy is considered when breakdown occurred. An experimental study is designed based on the proposed approach and solved by combinational genetic algorithm.

As chatbots progressively replace employees in customer support, studies on their influence on customer satisfaction, customers’ planning of the repurchasing, and customers’ recommendations of goods and services are rapidly expanding, assisting businesses in identifying implementable technical solutions that decrease labor expenses while preserving purchaser retention. However, the topic of influence of chatbots on the labor market mostly consists of sensationalist claims in the media about the disappearance of human job positions in customer service with the observable lack of rigorous empirical studies on the extent of job losses, AI-human collaboration models, and customer preferences. The findings of this study indicate that chatbot implementation has led to a moderate reduction in customer service staffing, with the largest effects in sectors characterized by routine inquiries. However, qualitative evidence shows that companies are mostly restructuring job roles rather than engaging in large scale layoffs, with many employees transitioning into more complex or strategic tasks. Customer satisfaction improved slightly, following digitalization and automation of customer service, suggesting that chatbots can enhance service efficiency when integrated into hybrid models. The Serbian context, particularly language adaptation challenges and uneven financial resources, and organizational readiness, continues to shape the pace and outcomes of implementation. These results highlight the need for targeted training and coordinated emplozee development policies to help employees in customer service adapt to changing job requirements.

Service composition platforms play a crucial role in creating personalized service processes. Challenges, including the risk of tampering with service data during service invocation and the potential single point of failure in centralized service registration centers, hinder the efficient and responsible creation of service processes. This paper presents a novel framework called Context-Aware Responsible Service Process Creation and Recommendation (SPCR-CA), which incorporates blockchain, Recurrent Neural Networks (RNNs), and a Skip-Gram model holistically to enhance the security, efficiency, and quality of service process creation and recommendation. Specifically, the blockchain establishes a trusted service provision environment, ensuring transparent and secure transactions between services and mitigating the risk of tampering. The RNN trains responsible service processes, contextualizing service components and producing coherent recommendations of linkage components. The Skip-Gram model trains responsible user-service process records, generating semantic vectors that facilitate the recommendation of similar service processes to users. Experiments using the Programmable-Web dataset demonstrate the superiority of the SPCR-CA framework to existing benchmarks in precision and recall. The proposed framework enhances the reliability, efficiency, and quality of service process creation and recommendation, enabling users to create responsible and tailored service processes. The SPCR-CA framework offers promising potential to provide users with secure and user-centric service creation and recommendation capabilities.

O2O (Online to Offline) is a new type of e-commerce model. Traditional recommendation methods focus on commodity rather than service, and lack the consideration on customer context and service status. In order to improve the quality of O2O service recommendation, this paper proposes an O2O service recommendation algorithm based on customer context and trust service. The method mainly includes two parts: (1) According to context sensitivity of service recommendation, a collaborative filtering recommendation algorithm is proposed which integrates basic user data and user context data, (2) In order to improve trust degree of service recommendation, the state QoS (Quality of Service) attributes of O2O service is updated continuously to ensure the availability of O2O service. Finally, two experiments are carried out based on the data set, and the experimental results show that the proposed algorithm has significant improvements in accuracy and trust degree of O2O service recommendation.

PurposeThe study is motivated by business' mixed response to increasing demand for customer service, leaving the question as to its impact on performance open. The study is concerned with the impact of customers' perception of customer service (bad/good) on variables that are known to drive revenue, i.e. customer satisfaction, perceived relative attractiveness, and commitment.Design/methodology/approachData were collected through a survey among bank customers. Two groups were sampled: customers who have experienced good or bad customer service. The hypotheses were tested by applying structural equation modeling and running two group analysis using the PLS and LISREL softwares.FindingsCustomers that experience bad customer service do take into account the same variables in their evaluation as do customers that experience good customer service. They do however, put different weights on every factor in the evaluation process. Also the strength of the relationships between the variables seems to differ. Typically, analyses showed that customers experiencing bad customer service tend to consider more thoroughly all aspects of the service; the relationships between the variables were stronger and the explained variance of each construct higher, than in the group of customers experiencing good customer service. However, the paths are not different across the groups.Research limitations/implicationsThe paper has only tested the model and hypotheses in one industry. Future research should test the same model using different industries reflecting different customer involvement levels.Practical implicationsFrom this study, service managers can learn that investing in customer service in ongoing customer relations is “the right thing to do” as it is linked to customer equity through customers' commitment to the firm. Second, as customer service in such relationships drives perceived relative attractiveness, saving the bottom line by cutting back on the human side of the customer interaction, may harm the firm's competitive position in the marketplace.Originality/valueThe impact of customer service on key performance variables in ongoing relations has to the knowledge never been studied before.