The Interplay Between Online Consumer Reviews and Recommender Systems: An Experimental Analysis

Baum, Daniela; Spann, Martin

doi:10.2753/jec1086-4415190104

Cited by 63 publications

(45 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…If comparing reviews and ratings, reviews might be considered more in-depth and insightful by providing additional information about a certain good (Tsekouras, 2017), however they require more effort to be processed (Tsekouras, 2017). Thus, retailers might support online consumers purchase decision by providing recommendations generated by both previous consumers and recommender systems (Baum and Spann, 2014). For instance, Amazon suggests new books to buy based on the past purchases of consumers with similar characteristics or on sophisticated algorithm (recommender systems) that try to match consumer interest with product offer.…”

Section: Online Consumer-generated Contentsmentioning

confidence: 99%

Making sense of consumers’ tweets

Pantano

Giglio

Dennis

2019

IJRDM

View full text Add to dashboard Cite

Purpose Consumers online interactions, posts, rating and ranking, reviews of products/attractions/restaurants and so on lead to a massive amount of data that marketers might access to improve the decision-making process, by impacting the competitive and marketing intelligence. The purpose of this paper is to develop understanding of consumers online-generated contents in terms of positive or negative comments to increase marketing intelligence. Design/methodology/approach The research focuses on the collection of 9,652 tweets referring to three fast fashion retailers of different sizes operating in the UK market, which have been shared among consumers and between consumer and firm, and subsequently evaluated through a sentiment analysis based on machine learning. Findings Findings provide the comparison and contrast of consumers’ response toward the different retailers, while providing useful guidelines to systematically making sense of consumers’ tweets and enhancing marketing intelligence. Practical implications The research provides an effective and systemic approach to accessing the rich data set on consumers’ experiences based the massive number of contents that consumers generate and share online and investigating this massive amount of data to achieve insights able to impact on retailers’ marketing intelligence. Originality/value To best of the authors’ knowledge, while other authors tried to identify the effect of positive or negative online comments/posts/reviews, the present study is the first one to show how to systematically detect the positive or negative sentiments of shared tweets for improving the marketing intelligence of fast fashion retailers.

show abstract

Section: Online Consumer-generated Contentsmentioning

confidence: 99%

Making sense of consumers’ tweets

Pantano

Giglio

Dennis

2019

IJRDM

View full text Add to dashboard Cite

show abstract

“…product's prior purchases and bookmarking, on consumers' purchase decisions. While few prior studies sparingly look into the interplay between multiple online information cues, they typically considered opinion-based cues such as online reviews or peer recommendations [7], [8], [12]. We contribute to the body of knowledge by introducing the interplay of action-based information cues.…”

Section: Theoretical Implicationsmentioning

confidence: 99%

“…Multiple information cues, when consistent across different sources, ease consumers' decision-making process by reinforcing each other. However, in reality consumers often face conflicting information about a product from different sources, challenging their decision making and weakening their inclination to purchase an item [7]. Recent studies investigate information processing in presence of inconsistent information and the heuristics followed by consumers to alleviate uncertainties raised by confounding information cues [8], [9].…”

Section: Introductionmentioning

confidence: 99%

Time-contingent Impact of Inconsistent Action-based Information Cues on Social Commerce Purchases: An Experimental Investigation

Bhattacharyya¹,

Bose²

2019

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

Social commerce websites predominantly display two types of action-based online social information: product's past purchases and bookmarks (e.g. wishlists). The impact of inconsistency between these two information cues on consumer decision making is uncertain and is expected to be dependent on the purchase context. In this paper we investigate the effect of (action-based) online social information inconsistency on consumers' likelihood of purchasing a product for temporally proximal and distant purchases. Using a controlled experimental setup with Latinsquare design and linear mixed model analysis we find significant interaction effect of information inconsistency type and temporal distance of purchase on purchase likelihood of product, establishing the purchase timing dependent impact of information inconsistency. The paper offers several academic implications, and valuable insights for website managers to elicit favorable consumer responses even under information inconsistency and effectively design their product recommendation strategies.

show abstract

“…There is evidence that two-sided reviews tend to be less trustworthy and helpful than onesided reviews [8,72,92]. This can be attributed to the fact that two-sided opinions elicit ambivalent feelings that are highly related to negative emotional outcomes, such as confusion and anxiety [70,83].…”

Section: Effects Of Online Reviews On Trials Of Mobile Appsmentioning

confidence: 99%