No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Conference Paper
Is a picture really worth a thousand words? - On the role of images in e-commerce
Abstract
In online peer-to-peer commerce places where physical examination of the goods is infeasible, textual descriptions, images of the products, reputation of the participants, play key roles. Visual image is a powerful channel to convey crucial information towards e-shoppers and influence their choice. In this paper, we investigate a well-known online marketplace where over millions of products change hands and most are described with the help of one or more images. We present a systematic data mining and knowledge discovery approach that aims to quantitatively dissect the role of images in e-commerce in great detail. Our goal is two-fold. First, we aim to get a thorough understanding of impact of images across various dimensions: product categories, user segments, conversion rate. We present quantitative evaluation of the influence of images and show how to leverage different image aspects, such as quantity and quality, to effectively raise sale. Second, we study interaction of image data with other selling dimensions by jointly modeling them with user behavior data. Results suggest that "watch" behavior encodes complex signals combining both attention and hesitation from buyer, in which image still holds an important role when compared to other selling variables, especially for products for which appearance is important. We conclude on how these findings can benefit sellers in a high competitive online e-commerce market.
... Nowadays, a common practice we can see is that customers upload pictures/videos of the purchased products and provide visual feedback [4], [5], besides (/instead of) the language feedback [3]. Such visual reviews can convey crucial implicit information regarding service and product quality, product handling during transit/delivery, user experience, etc., which C. Adak is with the Dept. of CSE, IIT Patna, India-801106, S. Chattopadhyay is with Dept. of CSE, IIIT Guwahati, India-781015, and Muhammad Saqib is with Data61, CSIRO, Australia-2122. ...
... Moreover, people often rely more on visual review than language one. Therefore, analyzing such visual reviews is becoming important [4]. ...
... ; e <t,t > = f A (r <t−1> , a <t > ); (4) where, f A is the alignment model, which is basically a feedforward neural network [14]. The decoder f D has total T y timesteps. ...
With the proliferation of the e-commerce industry, analyzing customer feedback is becoming indispensable to a service provider. In recent days, it can be noticed that customers upload the purchased product images with their review scores. In this paper, we undertake the task of analyzing such visual reviews, which is very new of its kind. In the past, the researchers worked on analyzing language feedback, but here we do not take any assistance from linguistic reviews that may be absent, since a recent trend can be observed where customers prefer to quickly upload the visual feedback instead of typing language feedback. We propose a hierarchical architecture, where the higher-level model engages in product categorization, and the lower-level model pays attention to predicting the review score from a customer-provided product image. We generated a database by procuring real visual product reviews, which was quite challenging. Our architecture obtained some promising results by performing extensive experiments on the employed database. The proposed hierarchical architecture attained a 57.48% performance improvement over the single-level best comparable architecture.
... Visual features play key roles in the purchasing process, which is possibly the most crucial factor in the decision making of a buyer [19]. Typically, a buyer will look at product pictures to obtain a general idea of the overall characteristics of the product before making his/her purchase decision. ...
... To demonstrate the impact of images in the prediction process, Di et al. [19] discussed the role of images in ecommerce by clarifying the impact of images across various dimensions. The authors stated that the "watch" behavior encoded complex signals in which the image played a key role compared to other selling variables. ...
Recent years have witnessed the rapid development of online shopping and ecommerce websites, e.g., eBay and OLX. Online shopping markets offer millions of products for sale each day. These products are categorized into many product categories. It is crucial for sellers to correctly estimate the price of the second-hand item. State-of-the-art methods can predict the price of only one item category. In addition, none of the existing methods utilized the price range of a given second-hand item in the prediction task, as there are several advertisements for the same product at different prices. In this vein, as the first contribution, we propose a deep model architecture for predicting the price of a second-hand item based on the image and textual description of the item for different sets of item types. This proposed method utilizes a deep neural network involving long short-term memory (LSTM) and convolutional neural network architectures for price prediction. The proposed model achieved a better mean absolute error accuracy score in comparison with the support vector machine baseline model. In addition, the second contribution includes twofold. First, we propose forecasting the minimum and maximum prices of the second-hand item. The models used for the forecasting task utilize linear regression, LSTM, and seasonal autoregressive integrated moving average methods. Second, we propose utilizing the model of the first contribution in predicting the item quality score. Then, the item quality score and the forecasted minimum and maximum prices are combined to provide the item’s final predicted price. Using a dataset crawled from a website for second-hand items, the proposed method of combining the predicted second-hand item quality score with the forecasted minimum and maximum price outperforms the other models in all of the used accuracy metrics with a significant performance gap.
... This line of work used hand-crafted image features, but did not actually assess the image quality as dependent variable. In another work on eCommerce, image quality was modelled and predicted through linear regression, and shown to be significant predictors of buyer interest [9]. However, the dataset is not available, nor any details on the modelling methodology or model performance. ...
... As shown in previous work [9], image quality matters more for product categories that are inherently more visual (e.g., clothing). Thus in our development of the dataset, we focus on the shoe and handbag categories. ...
As any savvy online shopper knows, second-hand peer-to-peer marketplaces are filled with images of mixed quality. How does image quality impact marketplace outcomes, and can quality be automatically predicted? In this work, we conducted a large-scale study on the quality of user-generated images in peer-to-peer marketplaces. By gathering a dataset of common second-hand products (~75,000 images) and annotating a subset with human-labeled quality judgments, we were able to model and predict image quality with decent accuracy (~87%). We then conducted two studies focused on understanding the relationship between these image quality scores and two marketplace outcomes: sales and perceived trustworthiness. We show that image quality is associated with higher likelihood that an item will be sold, though other factors such as view count were better predictors of sales. Nonetheless, we show that high quality user-generated images selected by our models outperform stock imagery in eliciting perceptions of trust from users. Our findings can inform the design of future marketplaces and guide potential sellers to take better product images.
... Data on sale ranks, list price, customer rating, number of reviewers, and days released from e-commerce sites could be used to forecast market demand, estimate cost and price elasticity, and even evaluate the optimality of pricing strategies [11]. Nowadays, not only texts but also images can be mined for competitors' products reputations [39]. Properties of images such as display formats, image quality, the number of views can affect buyers' intention, stimulate trust and improve the transaction rate [39], [40]. ...
... Nowadays, not only texts but also images can be mined for competitors' products reputations [39]. Properties of images such as display formats, image quality, the number of views can affect buyers' intention, stimulate trust and improve the transaction rate [39], [40]. ...
The digital transformation enables enterprises to mine big data for marketing intelligence on markets, customers, products, and competitor. However, there is a lack of a comprehensive literature review on this issue. With an aim to support enterprises to accelerate the digital transformation and gain competitive advantages through exploiting marketing intelligence from big data, this paper examines the literature in the period from 2001–2018. Consequently, 76 most relevant articles are analyzed based on four marketing intelligence components (Markets, Customers, Products, and Competitors) and six data mining models (Association, Classification, Clustering, Regression, Prediction, and Sequence Discovery). The findings of this study indicate that the research area of product and customer intelligence receives most research attention. This paper also provides a roadmap to guide future research on bridging marketing and information systems through the application of data mining to exploit marketing intelligence from big data.
... Data on sale ranks, list price, customer rating, number of reviewers, and days released from e-commerce sites could be used to forecast market demand, estimate cost and price elasticity, and even evaluate the optimality of pricing strategies [11]. Nowadays, not only texts but also images can be mined for competitors' products reputations [39]. Properties of images such as display formats, image quality, the number of views can affect buyers' intention, stimulate trust and improve the transaction rate [39], [40]. ...
... Nowadays, not only texts but also images can be mined for competitors' products reputations [39]. Properties of images such as display formats, image quality, the number of views can affect buyers' intention, stimulate trust and improve the transaction rate [39], [40]. ...
The digital transformation enables enterprises to mine big data for marketing intelligence on markets, customers, products, and competitor. However, there is a lack of a comprehensive literature review on this issue. With an aim to support enterprises to accelerate the digital transformation and gain competitive advantages through exploiting marketing intelligence from big data, this paper examines the literature in the period from 2001–2018. Consequently, 76 most relevant articles are analyzed based on four marketing intelligence components (Markets, Customers, Products, and Competitors) and six data mining models (Association, Classification, Clustering, Regression, Prediction, and Sequence Discovery). The findings of this study indicate that the research area of product and customer intelligence receives most research attention. This paper also provides a roadmap to guide future research on bridging marketing and information systems through the application of data mining to exploit marketing intelligence from big data.
... As discussed in [6], the major problem with approaches relying on textual features is the reliability of such features. One example are e-shops where people choose between many items based on their properties. ...
... There have been many studies that discuss the importance of visual stimuli on the user preferences. Di. et al. [6] reported that a product's salability could be improved by showing a user more product images with higher-quality. He et. ...
Recommender systems aim at enhancing user experience on the Web by employing the results of users behavior analysis for recommending items. However, user behavior is usually influenced by various aspects. Even though visual stimuli greatly influence almost every part of our life, it is yet poorly reflected in the domain of recommendation. In our work, we study the impact of visual stimuli (specifically images) on recommendation process on the Web. We focus on the domains where the impact of images is substantial (e.g., movies and shopping). First results of our experiments suggest that features extracted from the images are able to improve the ranking of the current recommendation approaches.
... A product description in a typical e-commerce marketplace is usually accompanied by a related image gallery. Whether those images are uploaded by the sellers themselves, or sourced by the e-commerce platform, they tend to considerably enrich user experience and provide customers with a valuable informational resource at various stages of their decision-making process [5,18]. ...
Image galleries provide a rich source of diverse information about a product which can be leveraged across many recommendation and retrieval applications. We study the problem of building a universal image gallery encoder through multi-task learning (MTL) approach and demonstrate that it is indeed a practical way to achieve generalizability of learned representations to new downstream tasks. Additionally, we analyze the relative predictive performance of MTL-trained solutions against optimal and substantially more expensive solutions, and find signals that MTL can be a useful mechanism to address sparsity in low-resource binary tasks.
... A study by Chen et al. [7] establishes a link between images and purchase intentions. The results presented by Di et al. [8] show positive evidence that better images can lead to increase in buyer's attention, trust and conversion rate. A recent study by Zakrewsky et al. [9] shows correlation between popular items and images with high quality images. ...
... Online sales currently limit the user to visual interaction with a product [57]. The lack of physical interaction between the buyer and seller may influence their interaction [58]. However, during this limited interaction one major component that contributes towards a product's successful purchase is visual product semantics [45]. ...
Assistive Technology (AT) product use occurs within a socio-cultural setting. The growth internationally in the AT product market suggests that designers need to be aware of the influences that diverse cultures may have on the societal perception of an AT product through its semantic attributes. The study aimed to evaluate the visual interaction with an AT product by young adults from Pakistan, a collectivist society, and the United Kingdom (UK), an individualist society. A paper-based questionnaire survey was carried out with 281 first-year undergraduate students from the UK and Pakistan to evaluate their perception towards the visual representation of a generic conventional wheelchair image. A semantics differential (SD) scale method was used involving a seven-point bipolar SD scale incorporating sixteen pairs of adjectives defining functional, meaning, and usability attributes of the product. The mean (M) and standard deviation (sd) values were obtained for each pair of adjectives and compared between both groups by employing appropriate parametric tests. The results show that having a diverse cultural background did not appear to have overtly influenced the meanings ascribed to the generic manual wheelchair, which was unexpected. The University ‘Internationalist’ environment may have influenced the results. Some minor but critical differences were found for some pairs of adjectives (bulky-compact, heavy-light), having p-value less than .05 (p < .05) that related to previous experience of wheelchairs and/or their use. Further studies are planned to investigate and validate outcomes with other student and non-student groups.
• Implications for Rehabilitation
• The semantic attributes of assistive technologies highlight a number of aspects that have implications for those involved in Assistive Technology (AT) product development, manufacturing and marketing.
• • For online sales, the AT products rely on the web page image to communicate the purpose and attributes of the product. There are limited explorations related to the semantic/communicative attributes of AT product presented in images, as perceived by individuals from diverse cultural backgrounds.
• • The knowledge towards semantic meaning ascribed to the AT product is important to investigate to provide a perspective that goes beyond practical functions of the AT product towards the communicative function.
• • Information of comprehending semantics and significance of the AT product from a social (non-users) viewpoint may benefits manufacturers in the development of AT products that best meet the societal needs, preferences and expectations.
• • A model of best practice, with a focus on semantic manipulation will offer Industrial Designers (ID) internationally with the suitable process and tools to reframe perceptions of disability and enhance acceptance of AT products not only for users, but also for the society around them.
... Also, item popularity highly depends on the image quality (Zakrewsky et al., 2016). (Di et al., 2014) provides deeper understanding of the roles images play in e-commerce and shows evidence that better images can lead to an increase of buyers' attention, trust, and conversion rates. ...
In e-commerce, product content, especially product images have a significant influence on a customer's journey from product discovery to evaluation and finally, purchase decision. Since many e-commerce retailers sell items from other third-party marketplace sellers besides their own, the content published by both internal and external content creators needs to be monitored and enriched, wherever possible. Despite guidelines and warnings, product listings that contain offensive and non-compliant images continue to enter catalogs. Offensive and non-compliant content can include a wide range of objects, logos, and banners conveying violent, sexually explicit, racist, or promotional messages. Such images can severely damage the customer experience, lead to legal issues, and erode the company brand. In this paper, we present a machine learning driven offensive and non-compliant image detection system for extremely large e-commerce catalogs. This system proactively detects and removes such content before they are published to the customer-facing website. This paper delves into the unique challenges of applying machine learning to real-world data from retail domain with hundreds of millions of product images. We demonstrate how we resolve the issue of non-compliant content that appears across tens of thousands of product categories. We also describe how we deal with the sheer variety in which each single non-compliant scenario appears. This paper showcases a number of practical yet unique approaches such as representative training data creation that are critical to solve an extremely rarely occurring problem. In summary, our system combines state-of-the-art image classification and object detection techniques, and fine tunes them with internal data to develop a solution customized for a massive, diverse, and constantly evolving product catalog.
... It is commonly formulated as rating prediction using matrix factorization [16]. It has been observed that images have a role in e-commerce [7,8]. When an item image is available, it could be used as additional feature. ...
Online reviews are prevalent. When recounting their experience with a product, service, or venue, in addition to textual narration, a reviewer frequently includes images as photographic record. While textual sentiment analysis has been widely studied, in this paper we are interested in visual sentiment analysis to infer whether a given image included as part of a review expresses the overall positive or negative sentiment of that review. Visual sentiment analysis can be formulated as image classification using deep learning methods such as Convolutional Neural Networks or CNN. However, we observe that the sentiment captured within an image may be affected by three factors: image factor, user factor, and item factor. Essentially, only the first factor had been taken into account by previous works on visual sentiment analysis. We develop item-oriented and user-oriented CNN that we hypothesize would better capture the interaction of image features with specific expressions of users or items. Experiments on images from restaurant reviews show these to be more effective at classifying the sentiments of review images.
... Clustering and association mining techniques are among the most common methods employed to support reputation management applications. More recently, Di et al.(2014) proposed a reputation management method which not only mines text-based reputation data from the Web but also considers the graphical images of products posted to the Web. Nevertheless, by the time of this writing, twenty billion images have been uploaded to Instagram. 1 Given such an extraordinary size of images archived online, it is extremely challenging to analyze the sheer volume of images for product reputation management, not to mention the variety of formats of source data (e.g., text versus images). ...
Big data analytics have been embraced as a disruptive technology that will reshape business intelligence, which is a domain that relies on data analytics to gain business insights for better decision-making. Rooted in the recent literature, we investigate the landscape of big data analytics through the lens of a marketing mix framework in this paper. We identify the data sources, methods, and applications related to five important marketing perspectives, namely people, product, place, price, and promotion, that lay the foundation for marketing intelligence. We then discuss several challenging research issues and future directions of research in big data analytics and marketing related business intelligence in general.
... The importance of using visual signals (i.e., product images) in recommendation scenarios has been stressed by previous works, e.g. [Di et al., 2014; Goswami et al., 2011] . State-of-the-art visuallyaware recommender systems make use of high-level visual features of product images extracted from (pre-trained) Deep CNNs on top of which a layer of parameterized transformation is learned to uncover those " visual dimensions " that predict users' actions (e.g. ...
Building successful recommender systems requires uncovering the underlying dimensions that describe the properties of items as well as users' preferences toward them. In domains like clothing recommendation, explaining users' preferences requires modeling the visual appearance of the items in question. This makes recommendation especially challenging, due to both the complexity and subtlety of people's 'visual preferences,' as well as the scale and dimensionality of the data and features involved. Ultimately, a successful model should be capable of capturing considerable variance across different categories and styles, while still modeling the commonalities explained by `global' structures in order to combat the sparsity (e.g. cold-start), variability, and scale of real-world datasets. Here, we address these challenges by building such structures to model the visual dimensions across different product categories. With a novel hierarchical embedding architecture, our method accounts for both high-level (colorfulness, darkness, etc.) and subtle (e.g. casualness) visual characteristics simultaneously.
... Users often make buying decisions purely by analyzing visual images and textual information available in print/digital advertisements, commercials and e-commerce websites [12][13]. As the popularity of e-commerce grows, a greater number of people every day are increasingly relying on visual representations of actual products for making buying decisions. ...
23 popular woodworking tool handle designs were collected to develop a reference visual catalog. Survey data was collected from 19 male craftsmen (11 carvers, 8 carpenters) regarding their most frequently used tools along with recommendations on handle designs suitable for these tools. Opinions were also collected regarding the most “liked” and “disliked” handle designs on the catalog. Subjective preferences and recommendations of handle designs for the most frequently used tools were also collected from a group of 58 undergraduate students of Design (41 male, 17 female). Participant responses revealed that woodcarvers most "liked” an ivory-colored, Japanese style, circular cross-section, regular-sized handle with a hooped top. Carpenters most "liked" a wenge-colored, rectangular-sectioned bulky handle with a hooped top. Male Design students most "liked" a golden-honey-colored, London pattern handle with an octagonal central section and domed wooden top. Female Design students most "liked" a beech-colored, bulky pear-shaped round handle with a hooped top. Overall, 12 different tools which included different sizes of u-gouges, v-parting tools, fishtail-chisels, firmer-chisels, mortise-chisels, and an in-denting tool were found to be the most frequently used implements by the craftsmen. On an aggregate for these 12 tools, an espresso-colored, bulky pear-shaped round handle with a hooped top was found to be the most recommended handle and a Bubinga-colored elongated pattern maker type handle was the least recommended handle. Results from this paper should help researchers and manufacturers gain qualitative insights into subjective preferences and biases that may exist for and against certain handle design features in the context of woodworking tool research and development
... For example, Xue and Muralidharan (2015) found that green visuals increase the positive assessments of environmental claims. Di et al. (2014) found that visual images increase buyers' attention, trust, and conversion rate. Therefore, we suggest that making the recycled components within a product visible to the customers, these images can influence reinforce the positive effects of using waste materials of customers directly into the products for that respective customer. ...
Our current take-make-dispose economic model faces a vital challenge as it extracts resources from the natural environment at faster rates than that the natural environment can replenish. A circular economy where businesses lower their negative impact on the natural environment by transitioning towards recycling business models (RBMs), one of the four principles of circularity, is suggested as a promising solution. For a RBM to become viable, collaboration among several stakeholders and across several industries is required. In addition, the RBM should be scalable to make a positive impact. Hence, developing RBMs is complex as organizations need to consider multiple principles imposed by the recycling, collaborative, and scalability dimensions of these business models (BMs). In addition, these principles often remain general and not actionable to the practitioners. Therefore, in this study, we researched the practical guidelines for viable RBMs that are also collaborative and scalable. The empirical setting is the reuse of textile fibers to develop biocomposite products. We studied three cases using a research-through-design approach. We contribute to the literature on RBMs by showing the six minimum practical guidelines for recyclability, collaboration, and scalability. We draw implications for within sector collaborations and advance the thought that lease constructs challenge the scalability of RBM.
... Extensive previous research have emphasized the importance of images in e-commerce scenarios (e.g. [6,10,11]). In recent years, there is a growing interest in investigating the visual compatibility between different items. ...
Building a successful recommender system depends on understanding both the dimensions of people's preferences as well as their dynamics. In certain domains, such as fashion, modeling such preferences can be incredibly difficult, due to the need to simultaneously model the visual appearance of products as well as their evolution over time. The subtle semantics and non-linear dynamics of fashion evolution raise unique challenges especially considering the sparsity and large scale of the underlying datasets. In this paper we build novel models for the One-Class Collaborative Filtering setting, where our goal is to estimate users' fashion-aware personalized ranking functions based on their past feedback. To uncover the complex and evolving visual factors that people consider when evaluating products, our method combines high-level visual features extracted from a deep convolutional neural network, users' past feedback, as well as evolving trends within the community. Experimentally we evaluate our method on two large real-world datasets from Amazon.com, where we show it to outperform state-of-the-art personalized ranking measures, and also use it to visualize the high-level fashion trends across the 11-year span of our dataset.
... Ürün görsellerindeki herhangi bir bulanıklık veya pikselleşme, ürünün kalitesiz olduğunu düşündürecek ve müşterileri uzaklaştıracaktır. Bunun sonucunda, e-ticaret sitelerinde yüksek kaliteli fotoğraflara sahip olmak, mutlak bir zorunluluktur ve müşterilerin satın almalarındaki karar verme sürecinde yardımcı olabilmektedir (Sean, 2019;Di, Sundaresan, Piramuthu ve Bhardwaj., 2014). Popüler olan "zoom" ve "kaydırma" özellikleri gibi çevrimiçi görselleştirme teknolojilerini benimsenmesi, ürün deneyimini kolaylaştırarak dönüşüm oranlarını artıracaktır (Song ve Kim, 2012). ...
Altyapı sağlayıcılar e-ticaret ekosisteminde köprü görevi görmektedir. E-ticaret ortamında iletişimin direkt olmaması bu köprüye daha
da önem kazandırmaktadır. Bu çalışmada, altyapı/arayüz sağlayıcı ve
yöneticileri perspektifinden e-ticaret alanında müşteri memnuniyetini
etkileyebilecek faktörler araştırılmaktadır. E-ticaret alanında müşteri
memnuniyetini etkileyen faktörlerle ilgili, sektördeki uzmanların görüşlerinin AHS yöntemi ile değerlendirilmesi yapılmıştır. Sonuçlara
göre Tedarik zincirinin hizmet sağlama halkasında yer alan uzmanlar
deneyimleri çerçevesinde öncelikli unsurun ürünün iyi fiyatlanmış ve
talebi karşılamaya yönelik olması gerektiği görüşünü ortaya koymuştur. Arayüz ikincil faktördür. Uzman görüşlerine göre müşteri ilişkilerinin rolü ise ancak bu iki faktörde e-ticaret başarılıysa etkinlik göstermektedir. Uzmanlar lojistik ile ilgili unsurlara sıralamada en sonda yer vermiştir
... To alleviate this problem, online fashion platforms offer various product information including categorical features such as fabric, occasion, length, etc. and visual imagery or video content in different setups. Studies have shown that product visual information (images/videos) plays a vital role in customer's confidence, click-through rate, and purchase decision [1,2]. However, due to the high costs in creating quality professional content, product images are often produced for only one single product size that flatters or best fits the fashion model who is wearing it. ...
Amidst the rapid growth of fashion e-commerce, remote fitting of fashion articles remains a complex and challenging problem and a main driver of customers' frustration. Despite the recent advances in 3D virtual try-on solutions, such approaches still remain limited to a very narrow - if not only a handful - selection of articles, and often for only one size of those fashion items. Other state-of-the-art approaches that aim to support customers find what fits them online mostly require a high level of customer engagement and privacy-sensitive data (such as height, weight, age, gender, belly shape, etc.), or alternatively need images of customers' bodies in tight clothing. They also often lack the ability to produce fit and shape aware visual guidance at scale, coming up short by simply advising which size to order that would best match a customer's physical body attributes, without providing any information on how the garment may fit and look. Contributing towards taking a leap forward and surpassing the limitations of current approaches, we present FitGAN, a generative adversarial model that explicitly accounts for garments' entangled size and fit characteristics of online fashion at scale. Conditioned on the fit and shape of the articles, our model learns disentangled item representations and generates realistic images reflecting the true fit and shape properties of fashion articles. Through experiments on real world data at scale, we demonstrate how our approach is capable of synthesizing visually realistic and diverse fits of fashion items and explore its ability to control fit and shape of images for thousands of online garments.
... Issues such as business intelligence, marketing intelligence, healthcare, security, and customer well-being are also tied to big data characteristics, which is why big data is interesting to researchers, businesspeople, and politicians. On the other hand, a large part of the big data related to click-through and relocation data through mobile devices requires high speed, which can be used for short-term forecasting with high accuracy [1][2][3][4][5][6][7][8][9][10][11][12][13][14] That's why big companies like Google and Facebook, which are skilled at analyzing huge amounts of data, are looking to build new businesses and explore big data. Manyika et. ...
Annotation: Big data has been increasingly considered by experts and company owners due to its direct impact on the development of businesses and companies, so that big data management increases the efficiency of important business decisions. With the increasing use of the Internet and the advancement of information storage technology, there is a vast amount of information that can be used and properly examined. This amount of information is called big data and can be used to make important decisions for the development of a company. All the while, many start-ups and even large corporations are unsure how to use big data. In addition to looking at big data and its role in decision making, this article examines how to properly extract information to support decision making.
... The importance of using visual signals (i.e., product images) in recommendation scenarios has been stressed by previous works, e.g. [Di et al., 2014; Goswami et al., 2011] . State-of-the-art visuallyaware recommender systems make use of high-level visual features of product images extracted from (pre-trained) Deep CNNs on top of which a layer of parameterized transformation is learned to uncover those " visual dimensions " that predict users' actions (e.g. ...
... Marketers and businesses constantly bombard social network users with e-commerce product images that hyperlink to store pages. This segment of online consumers is believed to rely disproportionately on these images as a source of product information and a basis to engage in purchasing decisions [3,4]. In fact, users of platforms rich in visual content including Tumblr, Instagram, and Facebook have started to adopt the trend of visual shopping [5][6][7]. ...
Background . Social network visual shopping trends are growing e-commerce at unprecedented levels. Images are used as product marketing material; however, image posts are triggering very low consumer behavior and low sales conversion. Objective . To explore how online stores can increase the purchasing prospects of their products using images on social networks. Methods . We introduce a theoretical probabilistic model to estimate consumer behavioral intention and purchasing prospect on social networks, outline parameters that can be exploited to increase click-rate and conversion, and motivate a new strategy to market products online. The model explores increasing online stores’ sales conversion by utilizing a product collection landing page that is marketed to consumers through a single “Hook” image. To implement the model, we developed a novel technological method that enabled online stores to post different “Hook” images on social networks and hyperlink them to the product collection landing pages they created. Results . Stores and marketers developed four types of “Hook” images: themed-collaged product images, single product images, lifestyle images, and model images. Themed-collaged product images accounted for 60% of consumer traffic from social network sites. Moreover, consumer purchasing click rate increased at least twofold (4.94%) with the use of product collection landing pages.
Recent study shows successful results in generating a proper language description for the given image, where the focus is on detecting and describing the contextual relationship in the image, such as the kind of object, relationship between two objects, or the action. In this paper, we turn our attention to more subjective components of descriptions that contain rich expressions to modify objects – namely attribute expressions. We start by collecting a large amount of product images from the online market site Etsy, and consider learning a language generation model using a popular combination of a convolutional neural network (CNN) and a recurrent neural network (RNN). Our Etsy dataset contains unique noise characteristics often arising in the online market. We first apply natural language processing techniques to extract high-quality, learnable examples in the real-world noisy data. We learn a generation model from product images with associated title descriptions, and examine how e-commerce specific meta-data and fine-tuning improve the generated expression. The experimental results suggest that we are able to learn from the noisy online data and produce a product description that is closer to a man-made description with possibly subjective attribute expressions.
Conversion is one of the ultimate goals of service providers. It requires an in-depth understanding of internal mental behavior toward conversion. There are many industrial best practices in terms of guidelines for conversion-oriented web design and smartphone application design. Most of them are still in a phase of trial-and-error because they lack a systematic methodology that deals with the mental flow of a customer during conversion. In order to provide building blocks towards a systematic construction of conversion-oriented web design methodology, the author proposes a 4-stage view model of conversion.
This chapter sheds light on various research areas of interest for further development of Mind Genomics. Research areas discussed refer to those of interest for the phase ONE of the process (avoiding the relatively high expenses related to market polling) and for the phase TWO of the process (avoiding relatively high expenses related to the estimation of the customer satisfaction level). Research of interest for the first phase is in media mining, while the research of interest for the second phase is in image understanding.
Perhaps the most exciting challenge and opportunity in entity retrieval is how to leverage entity-specific properties—attributes, types, and relationships—to improve retrieval performance. In this chapter, we take a departure from purely term-based approaches toward semantically enriched retrieval models. We look at a number of specific entity retrieval tasks that have been studied at various benchmarking campaigns. Specifically, these tasks are ad hoc entity retrieval, list search, related entity finding, and similar entity search. Additionally, we also consider measures of (static) entity importance.
In e-commerce, content quality of the product catalog plays a key role in delivering a satisfactory experience to the customers. In particular, visual content such as product images influences customers' engagement and purchase decisions. With the rapid growth of e-commerce and the advent of artificial intelligence, traditional content management systems are giving way to automated scalable systems. In this paper, we present a machine learning driven visual content management system for extremely large e-commerce catalogs. For a given product, the system aggregates images from various suppliers, understands and analyzes them to produce a superior image set with optimal image count and quality, and arranges them in an order tailored to the demands of the customers. The system makes use of an array of technologies, ranging from deep learning to traditional computer vision, at different stages of analysis. In this paper, we outline how the system works and discuss the unique challenges related to applying machine learning techniques to real-world data from e-commerce domain. We emphasize how we tune state-of-the-art image classification techniques to develop solutions custom made for a massive, diverse, and constantly evolving product catalog. We also provide the details of how we measure the system's impact on various customer engagement metrics.
Fotografi adalah teknik yang selalu ada dalam kehidupan kita, tidak jarang dalam teknik fotografi selalu menampilkan sisi yang realistik pada setiap visual. Korporat dan UMKM tidak jarang selalu menggunakan teknik fotografi pada medianya, namun selama ini strategi dan model visualnya selalu menggunakan teknik fotografi pada media. Urgensi pada penelitian ini yang menarik untuk dikaji dan di paparkan dalam artikel penelitian ini. Penelitian dengan judul “Strategi Visual Untuk Menyampaikan Citra Korporat Dan UMKM Ke Stakehokder Melalui Teknik Fotografi” mengambil studi kasus pada visual fotografi PT Swabina Gatra (anak perusahaan Semen Indonesia) dan UMKM Batik, CV Bintang Abadi. Penelitian ini bersifat deskriptif, dan mendukung pola merancang dengan teknik fotografi pada media. Pengumpulan data menggunakan metode pengumpulan portofolio dari kedua sampe diatas, selain itu penelitian ini melakukan wawancara, observasi dan literatur ilmiah.
The most labour-intensive stage of machine learning (ML) modelling is the appropriate preparation of correct dataset. This paper aims to show transfer dataset approach in image segmentation use case to lower labour intensity. Moreover, we test the effectiveness of this approach by training deep learning models on our prepared dataset. The models achieved high-performance metrics, even on very hard test data.
Previous studies have demonstrated that images are of great importance in attracting people’s attention and motivating them to take action. Various attributes (e.g., colors, aesthetics, and embedded objects) related to images are considered driving factors. Among which emotions in images, in particular, play a critical role in stimulating individuals, based on the Stimulus–Organism–Response theory. Consequently, many researchers put great efforts to understand image emotions, ranging from developing theoretical models to a broad spectrum of applications. Due to the complex and unstructured characteristics of images, identifying image emotions is challenging. Although some significant progress in image emotion classification has been achieved, inherent constraints still remain unaddressed. For example, acquiring a sufficiently large amount of labeled data to train a good model is costly and inevitably requires lots of human efforts. Besides, building a generalized model applicable to different datasets still needs a deep exploration. Image emotions are very subjective, which also makes such a classification task difficult. This paper proposes a general meta-learning framework for the few-shot image emotion classification, called Meta-IEC. Meta-IEC provides the capability of: (i) adapting to a similar dataset but new classes that have not been encountered before, and (ii) generalizing to a completely different dataset where emotion classes are unseen in the training dataset and only very few labeled images are available. Meta-IEC is also able to capture the uncertainty and ambiguity during the meta-testing, where we implement a hierarchical Bayesian graphical model to understand latent relationships among various parameters between meta-training and meta-testing. Extensive experiments on three commonly used datasets empirically demonstrate the superiority of our method over several state-of-the-art baselines. For example, our meta-learning based model can achieve performance improvement up to 5+%. We also provide some managerial implications on parameter sensitivity and label selection of meta-training and meta-testing.
Product images are essential for providing desirable user experience in an e-commerce platform. For a platform with billions of products, it is extremely time-costly and labor-expensive to manually pick and organize qualified images. Furthermore, there are the numerous and complicated image rules that a product image needs to comply in order to be generated/selected. To address these challenges, in this paper, we present a new learning framework in order to achieve Automatic Generation of Product-Image Sequence (AGPIS) in e-commerce. To this end, we propose a Multi-modality Unified Image-sequence Classifier (MUIsC), which is able to simultaneously detect all categories of rule violations through learning. MUIsC leverages textual review feedback as the additional training target and utilizes product textual description to provide extra semantic information. Based on offline evaluations, we show that the proposed MUIsC significantly outperforms various baselines. Besides MUIsC, we also integrate some other important modules in the proposed framework, such as primary image selection, noncompliant content detection, and image deduplication. With all these modules, our framework works effectively and efficiently in JD.com recommendation platform. By Dec 2021, our AGPIS framework has generated high-standard images for about 1.5 million products and achieves 13.6% in reject rate.
Fashion is an area that is in constant growth. The proliferation of social media and the Web, in general, has made e-shopping, thus corresponding recommender systems, increasingly important. Fashion recommender systems is a related area that we focus on in this chapter. More specifically, we present how recommender systems are used in online fashion stores to enhance the user experience and increase sales. In addition, we look at challenges the fashion domain specifically faces. We exemplify solution strategies by considering the SoBazaar system, including showing how we built a recommendation approach for the system and discussing results from our experiments. The results from these experiments demonstrate the effectiveness and viability of our method.
Product search engine faces unique challenges that differ from web page search. The goal of a product search engine is to rank relevant items that the user may be interested in purchasing. Clicks provide a strong signal of a user's interest in an item. Traditional click prediction models include many features such as document text, price, and user information. In this paper, we propose adding information extracted from the thumbnail image of the item as additional features for click prediction. Specifically, we use two types of image features -- photographic features and object features. Our experiments reveal that both types of features can be highly useful in click prediction. We measure our performance in both prediction accuracy and NDCG. Overall, our experiments show that augmenting with image features to a standard model in click prediction provides significant improvement in precision and recall and boosts NDCG.
Older consumers comprise a growing but under-represented segment of Internet users. However, compared to many younger groups, members of this segment often possess more discretionary time and income. This presents a significant opportunity for marketers of Internet related products and services. In order to better understand older individuals’ attitudes and motivations concerning Internet usage, phenomenological interviews were conducted among six Internet users and six non-users. From the emic perspective of the informants, and the etic interpretation of the transcripts, the following six themes characterizing differences between Internet using and Internet non-using older individuals emerged: Reference group affiliation, Technology schema, Resistance to change, Nature of social relations, Perception of reality, and Physical dexterity. The marketing implications of these findings are identified and discussed.
Consumers shop online for both goal-oriented and experiential reasons. However, goal-oriented motives are more common among online shoppers than are experiential motives. This article identifies and discusses attributes that facilitate goal-oriented online shopping, including accessibility/convenience, selection, information availability, and lack of unwanted sociality from retail sales help or shopping partners such as spouses. Importantly, consumers report that shopping online results in a substantially increased sense of freedom and control as compared to offline shopping. While consumers are more likely to describe offline rather than online shopping in experiential terms, evidence of experiential motivations for online shopping is emerging. Also, while closing transactions at web sites is one important e-commerce goal, companies should not lose site of the continuing importance and power of their web site as an information and communications vehicle.
While a large number of consumers in the US and Europe frequently shop on the Internet, research on what drives consumers to shop online has typically been fragmented. This paper therefore proposes a framework to increase researchers' understanding of consumers' attitudes toward online shopping and their intention to shop on the Internet. The framework uses the constructs of the Technology Acceptance Model (TAM) as a basis, extended by exogenous factors and applies it to the online shopping context. The review shows that attitudes toward online shopping and intention to shop online are not only affected by ease of use, usefulness, and enjoyment, but also by exogenous factors like consumer traits, situational factors, product characteristics, previous online shopping experiences, and trust in online shopping.
This research investigates eBay auction features that influence auction outcomes: likelihood of transaction or whether the item was actually sold in the auction (auction success) and value of the last bid (auction effectiveness). Specifically, we study several seller options available to the seller to increase the final price and successful end to the auction. We investigate the effectiveness of the term "New In Box" in the auction's heading, the use of an actual or cut picture, the initial price set by the seller, use of a reserve price and acceptance of a credit card in increasing the likelihood of the auction ending successfully, and at increasing the final price. Along with other independent variables, the impact of factors dealing with the auction pictures is examined. Hundreds of auctions (423) for two financial calculators were examined in this study. Findings show that utilization of certain risk-reducing auction features positively influence outcomes of these eBay auctions. These features include level of the starting bid, mention of "New in Box," inclusion of a real picture of the unit being sold, and the inclusion of a stock picture. In addition, findings indicate that certain risk-enhancing auction features negatively influence eBay auction outcomes. These features include the presence of a reserve price and the mention of "Wear" in the auction.
As use of the Internet has increased, many issues of trust have arisen. Users wonder: will my privacy be protected if I provide information to this Internet vendor? Will my credit card remain secure? Should I trust that this party will deliver the goods? Will the goods be as described? These questions are not merely academic. A recent Boston Consulting Group study revealed that one out of ten consumers have ordered and paid for items online that never were delivered (Williams, 2001). This year consumers filed around 11,000 complaints with the Federal Trade Commission alleging auction fraud, a figure up from the 107 lodged in 1997. It is no wonder that people are increasingly worried about whom to trust in online interactions. This paper explores the conditions under which online trust thrives and looks at examples of best and worst corporate practices. Online trust issues arise in a wide array of forums – chat rooms, news postings, e-catalogues, and retail transactions, to name a few. This paper focuses primarily on the online retail market, but the analysis applies to informational and entertainment sites as well.
Are trust and risk important in consumers' electronic commerce purchasing decisions? What are the antecedents of trust and risk in this context? How do trust and risk affect an Internet consumer's purchasing decision? To answer these questions, we i) develop a theoretical framework describing the trust-based decision-making process a consumer uses when making a purchase from a given site, ii) test the proposed model using a Structural Equation Modeling technique on Internet consumer purchasing behavior data collected via a Web survey, and iii) consider the implications of the model. The results of the study show that Internet consumers' trust and perceived risk have strong impacts on their purchasing decisions. Consumer disposition to trust, reputation, privacy concerns, security concerns, the information quality of the Website, and the company's reputation, have strong effects on Internet consumers' trust in the Website. Interestingly, the presence of a third-party seal did not strongly influence consumers' trust.
Web-based electronic stores have become more and more popular. However, there are not many guidelines, nor theories, showing what features of a store would work and why. This paper develops a set of functional guidelines for designing electronic stores and classifies them into three categories: motivational, hygiene, and media richness factors. An empirical study was conducted to evaluate the relative effect of these factors. The results show that the store design does have an effect on consumer purchase decision. A two-factor theory is plausible: hygiene factors are the major concern when consumers decide whether to shop electronically, while motivational factors play a key role when consumers choose among different electronic stores. Media richness factors are, in general, less important. The implication of the findings is that, for a web store to beat its electronic competitors, providing good transactional support is the key. If they would like to attract customers from traditional stores, special attentions must be paid to the hygiene factors.
The study of online consumer behavior is one of the most important research agendas in management information systems and marketing science. However, there is very limited knowledge about online consumer behavior because it is a complicated social-technical phenomenon and involves too many antecedent factors. Most prior studies in this area often offered inconsistent or even conflicted results due to using various simple research models in order to achieve parsimony. This study aims to overcome the drawback of previous studies and examines up to ten antecedent factors in one research model. ^ This study extends theory of planned behavior (TPB) by including ten important antecedents as external beliefs to online consumer behavior. The ten antecedents are identified by prior studies, mostly in the areas of management information systems and marketing science. This study is conducted with a survey of 288 college students who have online shopping experiences. The collected survey data is used to test each hypothesis developed in the research model. The results of data analysis confirm perceived ease of use (PEOU) and trust are essential antecedents in determining online consumer behavior through behavioral attitude and perceived behavioral control. The findings also indicate that cost reduction helps the consumer create positive attitude toward purchase. Further, the findings show the effects of two constructs of flow - concentration and telepresence on consumer’s attitude. Concentration is positively related to attitude toward purchase, but telepresence likely decreases attitude due to the consumer’s possible nervousness or concern about uncertainty in the online environment. ^ One of the main contributions of this study is that it provides valuable empirical evidence to prove that the TPB-based research model can well handle up to ten external beliefs with combining the partial least squares (PLS) statistical analysis. According to the findings, it is reasonable to conclude the TPB-based model has high potential to manipulate more than ten external beliefs without compromising the model’s goodness of fit.
A systems model is used to illustrate the information flow between three subsystems in e-commerce: Store Environment, Customer, and Web Technology. In the process of purchase, a customer makes several decisions: to enter the store, to navigate, to purchase, and to pay. This artificial environment must be designed so that it can support customer decision-making. At the same time it must be pleasing and fun, and create a task with natural flow. The paper summarizes existing research and theories, and provides a few suggestions for future research.
The purpose of this study is to analyze factors affecting on online shopping behavior of consumers that might be one of the most important issues of e-commerce and marketing field. However, there is very limited knowledge about online consumer behavior because it is a complicated socio-technical phenomenon and involves too many factors. One of the objectives of this study is covering the shortcomings of previous studies that didn't examine main factors that influence on online shopping behavior. This goal has been followed by using a model examining the impact of perceived risks, infrastructural variables and return policy on attitude toward online shopping behavior and subjective norms, perceived behavioral control, domain specific innovativeness and attitude on online shopping behavior as the hypotheses of study. To investigate these hypotheses 200 questionnaires dispersed among online stores of Iran. Respondents to the questionnaire were consumers of online stores in Iran which randomly selected. Finally regression analysis was used on data in order to test hypothesizes of study. This study can be considered as an applied research from purpose perspective and descriptive-survey with regard to the nature and method (type of correlation).
The study identified that financial risks and non-delivery risk negatively affected attitude toward online shopping. Results also indicated that domain specific innovativeness and subjective norms positively affect online shopping behavior. Furthermore, attitude toward online shopping positively affected online shopping behavior of consumers.
This research investigates eBay auction properties that influence auction outcomes: effectiveness (ability to attract bidders) and success (whether the item was actually sold in the auction). Hundreds of auctions (1,273) for three popular financial calculators—the HP 12c, the HP 10b and BA II plus—were examined to assess the impact of these auction properties. Findings show that utilization of certain auction features influence both outcomes. These features include starting price and reserve price for both calculators. Other features impact only auction effectiveness: wear and the presence of a picture for the HP10 and BA II plus; and wear and ability to pay with credit cards for the HP 12c. Further, others auction factors influence only auction success: abnormally high first bids and whether the seller is a recognized company rather than an individual for the HP 10b and BA II plus; and the length of time the seller has been selling on eBay, the presence of a picture, and whether the seller was a recognized company rather than an individual for the HP12c.
A model of demand for the Internet and other information sources is presented that treats the Internet as a production factor employed in producing benefits of search. Based upon the premise that the Internet is most efficient at providing information about functional attributes and price, several propositions are developed about its use and its impact on the use of other information sources. The model is supported by empirical evidence, using the example of Internet deployment in the search for a new automobile.
Although auctions have been examined extensively in economics, and to some degree in marketing, on-line auctions are only beginning to receive research attention. Further, in both economics and marketing the research on auctions has relied primarily on rational, economic theories. This article investigates how particular on-line auction features impact two important outcomes: auction success and final closing price. Traditional economic theories as well as theories from marketing and psychology are employed to provide a broader picture of on-line auctions. Specifically, several key factors related to auction success and closing price for four types of sterling flatware in an on-line auction site (eBay) are examined. The findings show that, across all four piece types, a reserve auction format, the relative opening price, and the number of bids unexplained by a low or high opening price are associated with both auction success and final closing price. © 2003 Wiley Periodicals, Inc.
Since George A. Akerlof (1970), economists have understood the adverse selection problem that information asymmetries can create in used goods mar-kets. The remarkable growth in online used goods auctions thus poses a puzzle. Part of the solution is that sellers voluntarily disclose their private information on the auction webpage. This defines a precise contract — to deliver the car shown for the closing price — which helps protect the buyer from adverse se-lection. I test this theory using data from eBay Motors, finding that online disclosures are important price determinants; and that disclosure costs impact both the level of disclosure and prices.
The Internet interface poses a difficulty for buyers in evaluating products online, particularly physical experience and durable goods, such as used cars. This increases buyers' product uncertainty, defined as the buyer's perceived estimate of the variance in product quality based on subjective probabilities about the product's characteristics and whether the product will perform as expected. However, the literature has largely ignored product uncertainty and mostly focused on mitigating buyer's seller uncertainty. To address this void, this study aims to conceptualize the construct of product uncertainty and propose its antecedents and consequences in online auction marketplaces. First, drawing upon the theory of markets with asymmetric information, we propose product uncertainty to be distinct from, yet affected by, seller uncertainty. Second, based on auction pricing theory, we propose that product uncertainty and seller uncertainty negatively affect two key success outcomes of online marketplaces: price premium and transaction activity. Third, following information signaling theory, we propose a set of product information signals to mitigate product uncertainty: (1) online product descriptions (textual, visual, multimedia); (2) third-party product certifications (inspection, history report, warranty); (3) auction posted prices (reserve, starting, buy-it-now); and (4) intrinsic product characteristics (book value and usage). Finally, we propose that the effect of online product descriptions and intrinsic product characteristics on product uncertainty is moderated by seller uncertainty. The proposed model is supported by a unique dataset comprised of a combination of primary (survey) data drawn from 331 buyers who bid upon a used car on eBay Motors, matched with secondary transaction data from the corresponding online auctions. The results distinguish between product and seller uncertainty, show the stronger role of product uncertainty on price premiums and transaction activity compared to seller uncertainty, empirically identify the most influential product information signals, and support the mediating role of product uncertainty. This paper contributes to and has implications for better understanding the nature and role of product uncertainty, identifying mechanisms for mitigating product uncertainty, and demonstrating complementarities between product and seller information signals. The model's generalizability and implications are discussed.
In this paper we study the importance of image based features on the click-through rate (CTR) in the context of a large scale product search engine. Typically product search engines use text based features in their ranking function. We present a novel idea of using image based features, common in the photography literature, in addition to text based features. We used a stochastic gradient boosting based regression model to learn relationships between features and CTR. Our results indicate statistically significant correlations between the image features and CTR. We also see improvements to NDCG and mean standard regression.
For buyers and sellers alike, there's no better way to earn one another's trust in online interactions.
yin comparing multiple products on the samescreen all have adverse effects on electronic shopping[2]. Can customers find what they want in the stores?Are customers aware of what products are available?After all, diligence in browsing a store is not a virtueretailers should expect from its online customers.We review online retail store attributes such as thenumber of links into the store, image sizes, numberof products, and store navigation features. By reviewingthe user...
This paper introduces a model for analyzing marketplaces, such as eBay, which rely on binary reputation mechanisms for quality signaling and quality control. In our model sellers keep their actual quality private and choose what quality to advertise. The reputation mechanism is primarily used to determine whether sellers advertise truthfully. Buyers may exercise some leniency when rating sellers, which needs to be compensated by corresponding strictness when judging sellers' feedback profiles. It is shown that, the more lenient buyers are when rating sellers, the more likely it is that sellers will find it optimal to settle down to steady-state quality levels, as opposed to oscillating between good quality and bad quality. Furthermore, the fairness of the market outcome is determined by the relationship between rating leniency and strictness when assessing a seller's feedback profile. If buyers judge sellers too strictly (relative to how leniently they rate) then, at steady state, sellers will be forced to understate their true quality. On the other hand, if buyers judge too leniently then sellers can get away with consistently overstating their true quality. An optimal judgment rule, which results in outcomes where at steady state buyers accurately estimate the true quality of sellers, is analytically derived. However, it is argued that this optimal rule depends on several system parameters, which are difficult to estimate from the information that marketplaces, such as eBay, currently make available to their members. It is therefore questionable to what extent unsophisticated buyers are capable of deriving and applying it correctly in actual settings.
Online auctions have recently gained widespread popularity and are one of the most successful forms of electronic commerce. We examine a dataset of eBay coin auctions to explore features of online bidding and selling behavior. We address three main issues. First, we measure the extent of the winner's curse. We find that for a representative auction in our sample, a bidder's expected profits fall by 3.2 percent when the expected number of bidders increases by one. Second, we document that costly entry is a key component in understanding observed bidding behavior. For a representative auction in our sample, a bidder requires $3.20 of expected profit to enter the auction. Third, we study the seller's choice of reserve prices. We find that items with higher book value tend to be sold using a secret as opposed to posted reserve price with a low minimum bid. We find that this is, to a first approximation, consistent with maximizing behavior. We also develop new techniques for structurally es...
We propose an analytical framework for studying bidding behavior in online auctions. The framework focuses on three key dimensions: the multi-stage process, the types of value-signals employed at each phase, and the dynamics of bidding behavior whereby early choices impact subsequent bidding decisions. We outline a series of propositions relating to the auction entry decision, bidding decisions during the auction, and bidding behavior at the end of an auction. In addition, we present the results of three preliminary field studies that investigate factors that influence consumers' value assessments and bidding decisions. In particular, (a) due to a focus on the narrow auction context, consumers under-search and, consequently, overpay for widely available commodities (CDs, DVDs) and (b) auction starting prices lead to higher winning bids, but only when comparable items are not available in the immediate context. We discuss the implications of this research with respect to our understanding of the key determinants of consumer behavior in this increasingly important arena of purchase decisions. - 2 - Web-based auctions have become one of the greatest successes of the Internet, success that has not diminished even after many other web-based services have lost their initial popularity. The growing importance of online auctions has attracted the attention of consumer researchers, who have studied such issues as herding behavior (Dholakia & Soltysinski, 2001), the impact of reserve prices (Hubl & Popkowski Leszczyc, 2001), the role of expertise (Wilcox, 2000), and the effects of auction formats (Lucking-Reiley, 1999). Still, our understanding of buyer (bidder) behavior in online auctions is rather limited. In particular, acquiring an item through online auctions is different i...
Online auctions: User experience insights from ebay
- A Haywood
Antecedents of desirable consumer behaviors in electronic commerce
- N Sukpanich
- L Chen
An investigation of university students' on-line shopping behavior
- H.-J Han
- R Ocker
- J Fjermestad
User experience insights from ebay
- Haywood A.