Search for: All records

Recommender systems traditionally find the most relevant products or services for users tailored to their needs or interests but they ignore the interests of the other sides of the market (aka stakeholders). In this paper, we propose to use a Ranked Bandit approach for an online multi-stakeholder recommender system that sequentially selects top 𝑘 items according to the relevance and priority of all the involved stakeholders. We presented three different criteria to consider the priority of each stakeholder when evaluating our approach. Our extensive experimental results on a movie dataset showed that the contextual multi-armed bandits with a relevance function make a higher level of satisfaction for all involved stakeholders in the long term. Keywords: Multi-stakeholder Recommender Systems; Multi-armed Bandits; Ranked Bandit; 

Traditional recommender systems help users find the most relevant products or services to match their needs and preferences. However, they overlook the preferences of other sides of the market (aka stakeholders) involved in the system. In this paper, we propose to use contextual bandit algorithms in multi-stakeholder platforms where a multi-sided relevance function with adjusting weights is modeled to consider the preferences of all involved stakeholders. This algorithm sequentially recommends the items based on the contextual features of users along with the priority of the stakeholders and their relevance to the items.Our extensive experimental results on a dataset consisting of MovieLens (1m), IMDB (81k+), and a synthetic dataset show that our proposed approach outperforms the baseline methods and provides a good trade-off between the satisfaction of different stakeholders over time. 

Recommender systems are widely used to help customers find the most relevant and personalized products or services tailored to their preferences. However, traditional systems ignore the preferences of the other side of the market, e.g., “product suppliers” or “service providers”, towards their customers. In this paper, we present 2SRS a Two-Sided Recommender System that recommends coupons, supplied by local businesses, to passerby while considering the preferences of both sides towards each other. For example, some passerby may only be interested in coffee shops whereas certain businesses may only be interested in sending coupons to new customers only. Our experimental results show that 2SRS delivers higher satisfaction when considering both sides of the market compared to the baseline methods. 

Public transit is one of the first things that come to mind when someone talks about “smart cities.” As a result, many technologies, applications, and infrastructure have already been deployed to bring the promise of the smart city to public transportation. Most of these have focused on answering the question, “When will my bus arrive?”; little has been done to answer the question, “How full will my next bus be?” which also dramatically affects commuters’ quality of life. In this article, we consider the bus fullness problem. In particular, we propose two different formulations of the problem, develop multiple predictive models, and evaluate their accuracy using data from the Pittsburgh region. Our predictive models consistently outperform the baselines (by up to 8 times). 

The pervasiveness of public displays is prompting an increased need for “fresh” content to be shown, that is highly engaging and useful to passerbys. As such, live or time-sensitive content is often shown in conjunction with “traditional” static content, which creates scheduling challenges. In this work, we propose a utility-based framework that can be used to represent the usefulness of a content item over time. We develop a novel scheduling algorithm for handling live and non-live content on public displays using our utility-based framework. We experimentally evaluate our proposed algorithm against a number of alternatives under a variety of workloads; the results show that our algorithm performs well on the proposed metrics. Additional experimental evaluation shows that our utility-based framework can handle changes in priorities and deadlines of content items, without requiring any involvement by the display owner beyond the initial setup. 

In this paper, we describe the implementation of an information sharing platform, got-toilet-paper.com. We create this web page in response to the COVID-19 pandemic to help the Pittsburgh, PA community share information about congestion and product shortages in supermarkets. We show that the public good problem of the platform makes it difficult for the platform to operate. In particular, there is sizable demand for the information, but supply satis es only a small fraction of demand. We provide a theoretical model and show that the first best outcomes cannot be obtained in a free market and the best symmetric equilibrium outcome decreases as the number of participant increases. Also, the best symmetric equilibrium has two problems, cost inefficiency and positive probability of termination. We discuss two potential solutions. The first is a uniform random sharing mechanism, which implies randomly selecting one person every period who will be responsible for information sharing. It is ex-post individually rational but hard to implement. The second solution is the one that we began implementing. It implies selecting a person at the beginning and make her responsible to share information every period, while reimbursing her cost. We discuss the reasons for high demand and low supply both qualitatively and quantitatively. 

The pervasiveness of public displays is prompting an increased need for "fresh" content to be shown, that is highly engaging and useful to passerbys. As such, live or time-sensitive content is often shown in conjunction with "traditional" static content, which creates scheduling challenges. In this work, we propose a utility-based framework and a novel scheduling algorithm for handling live and non-live content on public displays. We also experimentally evaluate our proposed algorithm against a number of alternatives under a variety of workloads. 

During recent years there have been several efforts from city and transportation planners, as well as, port authorities, to design multimodal transport systems, covering the needs of the population to be served. However, before designing such a system, the first step is to understand the current gaps. Does the current system meet the transit demand of the geographic area covered? If not, where are the gaps between supply and demand? To answer this question, the notion of transit desert has been introduced. A transit desert is an area where the supply of transit service does not meet the demand for it. While there is little ambiguity on what constitutes transit demand, things are more vague when it comes to transit supply. Existing efforts often define transit supply using volume metrics (e.g., number of bus stops within a pre-defined distance). However, this does not necessarily capture the quality of the transit service. In this study, we introduce a network-based transit desert index (which we call TDI) that captures not only the quantity of transit supply in an area, but also the connectivity that the transit system provides for an area within the region of interest. In particular, we define a network between areas based on the transit travel time, distance, and overall quantity of connections. We use these measures to examine two notions of transit quality: connectivity and availability. To quantify the connectivity of an area i we utilize the change observed in the second smallest eigenvalue of the Laplacian when we remove node i from the network. To quantify availability of an area i, we examine the number of routes which pass through this area as given by an underlying transit network. We further apply and showcase our approach with data from Allegheny County, Pennsylvania, USA. Finally, we discuss current limitations of TDI and how we can tackle them as part of our future research.