skip to main content

Title: 2SRS: A Two-Sided Recommender System to Connect Local Businesses to Bus Passengers
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.
Authors:
;
Award ID(s):
1739413
Publication Date:
NSF-PAR ID:
10298332
Journal Name:
22nd IEEE International Conference on Mobile Data Management (MDM 2021)
Page Range or eLocation-ID:
127 to 132
Sponsoring Org:
National Science Foundation
More Like this
  1. A large number of two-sided markets are now mediated by search and recommender systems, ranging from online retail and streaming entertainment to employment and romantic-partner matching. I will discuss in this talk how the design decisions that go into these search and recommender systems carry substantial power in shaping markets and allocating opportunity to the participants. This does not only raise legal and fairness questions, but also questions about how these systems shape incentives and the long-term effectiveness of the market. At the core of these questions lies the problem of where to rank each item, and how this affectsmore »both sides of the market. While it is well understood how to maximize the utility to the users, this talk focuses on how rankings affect the items that are being ranked. From the items perspective, the ranking system is an arbiter of exposure and thus economic opportunity. I will discuss how machine learning algorithms that follow the conventional Probability Ranking Principle [1] can lead to undesirable and unfair exposure allocation for both exogenous and endogenous reasons. Exogenous reasons often manifest themselves as biases in the training data, which then get reflected in the learned ranking policy. But even when trained with unbiased data, reasons endogenous to the system can lead to unfair or undesirable allocation of opportunity. To overcome these challenges, I will present new machine learning algorithms [2,3,4] that directly address both endogenous and exogenous factors, allowing the designer to tailor the ranking policy to be appropriate for the specific two-sided market.« less
  2. Context has been recognized as an important factor to consider in personalized recommender systems. Particularly in location-based services (LBSs), a fundamental task is to recommend to a mobile user where he/she could be interested to visit next at the right time. Additionally, location-based social networks (LBSNs) allow users to share location-embedded information with friends who often co-occur in the same or nearby points-of-interest (POIs) or share similar POI visiting histories, due to the social homophily theory and Tobler’s first law of geography. So, both the time information and LBSN friendship relations should be utilized for POI recommendation. Tensor completion hasmore »recently gained some attention in time-aware recommender systems. The problem decomposes a user-item-time tensor into low-rank embedding matrices of users, items and times using its observed entries, so that the underlying low-rank subspace structure can be tracked to fill the missing entries for time-aware recommendation. However, these tensor completion methods ignore the social-spatial context information available in LBSNs, which is important for POI recommendation since people tend to share their preferences with their friends, and near things are more related than distant things. In this paper, we utilize the side information of social networks and POI locations to enhance the tensor completion model paradigm for more effective time-aware POI recommendation. Specifically, we propose a regularization loss head based on a novel social Hausdorff distance function to optimize the reconstructed tensor. We also quantify the popularity of different POIs with location entropy to prevent very popular POIs from being over-represented hence suppressing the appearance of other more diverse POIs. To address the sensitivity of negative sampling, we train the model on the whole data by treating all unlabeled entries in the observed tensor as negative, and rewriting the loss function in a smart way to reduce the computational cost. Through extensive experiments on real datasets, we demonstrate the superiority of our model over state-of-the-art tensor completion methods.« less
  3. Problem definition : Participants in matching markets face search and screening costs when seeking a match. We study how platform design can reduce the effort required to find a suitable partner. Practical/academic relevance : The success of matching platforms requires designs that minimize search effort and facilitate efficient market clearing. Methodology : We study a game-theoretic model in which “applicants” and “employers” pay costs to search and screen. An important feature of our model is that both sides may waste effort: Some applications are never screened, and employers screen applicants who may have already matched. We prove existence and uniquenessmore »of equilibrium and characterize welfare for participants on both sides of the market. Results : We identify that the market operates in one of two regimes: It is either screening-limited or application-limited. In screening-limited markets, employer welfare is low, and some employers choose not to participate. This occurs when application costs are low and there are enough employers that most applicants match, implying that many screened applicants are unavailable. In application-limited markets, applicants face a “tragedy of the commons” and send many applications that are never read. The resulting inefficiency is worst when there is a shortage of employers. We show that simple interventions—such as limiting the number of applications that an individual can send, making it more costly to apply, or setting an appropriate market-wide wage—can significantly improve the welfare of agents on one or both sides of the market. Managerial implications : Our results suggest that platforms cannot focus exclusively on attracting participants and making it easy to contact potential match partners. A good user experience requires that participants not waste effort considering possibilities that are unlikely to be available. The operational interventions we study alleviate congestion by ensuring that potential match partners are likely to be available.« less
  4. Abstract Professional and lifelong learning are a necessity for workers. This is true both for re-skilling from disappearing jobs, as well as for staying current within a professional domain. AI-enabled scaffolding and just-in-time and situated learning in the workplace offer a new frontier for future impact of AIED. The hallmark of this community’s work has been i) data-driven design of learning technology and ii) machine-learning enabled personalized interventions. In both cases, data are the foundation of AIED research and data-related ethics are thus central to AIED research. In this paper we formulate a vision how AIED research could address data-relatedmore »ethics issues in informal and situated professional learning. The foundation of our vision is a secondary analysis of five research cases that offer insights related to data-driven adaptive technologies for informal professional learning. We describe the encountered data-related ethics issues. In our interpretation, we have developed three themes: Firstly, in informal and situated professional learning, relevant data about professional learning – to be used as a basis for learning analytics and reflection or as a basis for adaptive systems - is not only about learners. Instead, due to the situatedness of learning, relevant data is also about others (colleagues, customers, clients) and other objects from the learner’s context. Such data may be private, proprietary, or both. Secondly, manual tracking comes with high learner control over data. Thirdly, learning is not necessarily a shared goal in informal professional learning settings. From an ethics perspective, this is particularly problematic as much data that would be relevant for use within learning technologies hasn’t been collected for the purposes of learning. These three themes translate into challenges for AIED research that need to be addressed in order to successfully investigate and develop AIED technology for informal and situated professional learning. As an outlook of this paper, we connect these challenges to ongoing research directions within AIED – natural language processing, socio-technical design, and scenario-based data collection - that might be leveraged and aimed towards addressing data-related ethics challenges.« less
  5. Suweis, Samir (Ed.)
    Statistical network models have been used to study the competition among different products and how product attributes influence customer decisions. However, in existing research using network-based approaches, product competition has been viewed as binary (i.e., whether a relationship exists or not), while in reality, the competition strength may vary among products. In this paper, we model the strength of the product competition by employing a statistical network model, with an emphasis on how product attributes affect which products are considered together and which products are ultimately purchased by customers. We first demonstrate how customers’ considerations and choices can be aggregatedmore »as weighted networks. Then, we propose a weighted network modeling approach by extending the valued exponential random graph model to investigate the effects of product features and network structures on product competition relations. The approach that consists of model construction, interpretation, and validation is presented in a step-by-step procedure. Our findings suggest that the weighted network model outperforms commonly used binary network baselines in predicting product competition as well as market share. Also, traditionally when using binary network models to study product competitions and depending on the cutoff values chosen to binarize a network, the resulting estimated customer preferences can be inconsistent. Such inconsistency in interpreting customer preferences is a downside of binary network models but can be well addressed by the proposed weighted network model. Lastly, this paper is the first attempt to study customers’ purchase preferences (i.e., aggregated choice decisions) and car competition (i.e., customers’ co-consideration decisions) together using weighted directed networks.« less