Improving daily deals recommendation using explore-then-exploit strategies

Abstract

Daily-Deals Sites (DDSs) enable local businesses, such as restaurants and stores, to promote their products and services and to increase their sales by offering customers significantly reduced prices. If a customer finds a relevant deal in the catalog of electronic coupons, she can purchase it and the DDS receives a commission. Thus, offering relevant deals to customers maximizes the profitability of the DDS. An immediate strategy, therefore, would be to apply existing recommendation algorithms to suggest deals that are potentially relevant to specific customers, enabling more appealing, effective and personalized catalogs. However, this strategy may be innocuous because (1) most of the customers are sporadic bargain hunters, and thus past preference data is extremely sparse, (2) deals have a short living period, and thus data is extremely volatile, and (3) customers' taste and interest may undergo temporal drifts. In order to address such a particularly challenging scenario, we propose a new algorithm for daily deals recommendation based on an explore-then-exploit strategy. Basically, we choose a fraction of the customers to gather feedback on the current catalog in the exploration phase, and the remaining customers to receive improved recommendations based on the previously gathered feedback in a posterior exploitation phase. During exploration, a co-purchase network structure is updated with customer feedback (i.e., the purchases of the day), and during exploitation the updated network is used to enrich the recommendation algorithm. An advantage of our approach is that it is agnostic to the underlying recommender algorithm. Using real data obtained from a large DDS in Brazil, we show that the way in which we split customers into exploration and exploitation impacts by large the effectiveness of the recommendations. We evaluate different splitting strategies based on network centrality metrics and show that our approach offers gains in mean average precision and mean reciprocal rank ranging from 14 to 34 % when applied on top of state-of-the-art recommendation algorithms.

Information and data management at PUC-rio and UFMG

“…Daily-Deals sites enable local businesses, such as restaurants and stores, to promote their products and services to increase their sales by offering customers significantly reduced prices. In [91], we propose a new algorithm for daily deals recommendation based on an explore-then-exploit strategy.…”

A Multi-Armed Bandit Model Selection for Cold-Start User Recommendation

“…Several works model the recommendation problem using a MAB se ing in which the items to be recommended are the arms [7,21,42]. In a di erent way, Lacerda et al model users as arms to recommend daily-deals [33,34]. ey consider strategies for spli ing users into exploration and exploitation.…”

Groupon and Groupon Now: Participating Firm’s Profitability Analysis

“…[Edelman et al 2014] has reviewed the related literature up to early 2014. More recent theoretical work include the signalling effect of daily deal promotion [Zhao et al 2014], auction mechanism design for daily-deal sites' revenue maximization [Chen et al 2015a], real-time advertisement bidding optimization for firms' revenue maximization [Balakrishnan and Bhatt 2015], information diffusion and sales prediction [Zhou et al 2013], threshold discounting effects and optimal design [Marinesi et al 2013;, price competition [Meir et al 2014], group-buying auctions [Chen and Chung 2015], group recommendation [Lacerda et al 2015], group formation [Roy et al 2015], firm optimal strategies for group-buying [Jiang and Deng 2014;Ni et al 2015], and bundle discounts [Coviello and Franceschetti 2015]. Recent empirical work include relationship between deal size, advertising effect, and firm's revenue [Bai et al 2015], and factors affecting consumer benefits creation [Chen et al 2015b;Yeh et al 2014].…”