Preprint

Understanding Image Quality and Trust in Peer-to-Peer Marketplaces

Authors:
Preprints and early-stage research may not have been peer reviewed yet.
To read the file of this research, you can request a copy directly from the authors.

Abstract

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.

No file available

Request Full-text Paper PDF

To read the file of this research,
you can request a copy directly from the authors.

ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
A number of online marketplaces enable customers to buy or sell used products, which raises the need for ranking tools to help them find desirable items among a huge pool of choices. To the best of our knowledge, no prior work in the literature has investigated the task of used product ranking which has its unique characteristics compared with regular product ranking. While there exist a few ranking metrics (e.g., price, conversion probability) that measure the “goodness” of a product, they do not consider the time factor, which is crucial in used product trading due to the fact that each used product is often unique while new products are usually abundant in supply or quantity. In this paper, we introduce a novel time-aware metric—“sellability”, which is defined as the time duration for a used item to be traded, to quantify the value of it. In order to estimate the “sellability” values for newly generated used products and to present users with a ranked list of the most relevant results, we propose a combined Poisson regression and listwise ranking model. The model has a good property in fitting the distribution of “sellability”. In addition, the model is designed to optimize loss functions for regression and ranking simultaneously, which is different from previous approaches that are conventionally learned with a single cost function, i.e., regression or ranking. We evaluate our approach in the domain of used vehicles. Experimental results show that the proposed model can improve both regression and ranking performance compared with non-machine learning and machine learning baselines.
Conference Paper
Full-text available
We present a method for detecting objects in images using a single deep neural network. Our approach, named SSD, discretizes the output space of bounding boxes into a set of default boxes over different aspect ratios and scales per feature map location. At prediction time, the network generates scores for the presence of each object category in each default box and produces adjustments to the box to better match the object shape. Additionally, the network combines predictions from multiple feature maps with different resolutions to naturally handle objects of various sizes. SSD is simple relative to methods that require object proposals because it completely eliminates proposal generation and subsequent pixel or feature resampling stages and encapsulates all computation in a single network. This makes SSD easy to train and straightforward to integrate into systems that require a detection component. Experimental results on the PASCAL VOC, COCO, and ILSVRC datasets confirm that SSD has competitive accuracy to methods that utilize an additional object proposal step and is much faster, while providing a unified framework for both training and inference. For \(300 \times 300\) input, SSD achieves 74.3 % mAP on VOC2007 test at 59 FPS on a Nvidia Titan X and for \(512 \times 512\) input, SSD achieves 76.9 % mAP, outperforming a comparable state of the art Faster R-CNN model. Compared to other single stage methods, SSD has much better accuracy even with a smaller input image size. Code is available at https:// github. com/ weiliu89/ caffe/ tree/ ssd.
Article
Full-text available
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.
Conference Paper
Full-text available
We present an interactive application that enables users to improve the visual aesthetics of their digital photographs using spatial re- composition. Unlike earlier work that focuses either on photo qual- ity assessment or interactive tools for photo editing, we enable the user to make informed decisions about improving the composition of a photograph and to implement them in a single framework. Specifically, the user interactively selects a foreground object and the system presents recommendations for where it can be moved in a manner that optimizes a learned aesthetic metric while obeying semantic constraints. For photographic compositions that lack a distinct foreground object, our tool provides the user with cropping or expanding recommendations that improve its aesthetic quality. We learn a support vector regression model for capturing image aesthetics from user data and seek to optimize this metric during recomposition. Rather than prescribing a fully-automated solution, we allow user-guided object segmentation and inpainting to ensure that the final photograph matches the user's criteria. Our approach achieves 86% accuracy in predicting the attractiveness of unrated images, when compared to their respective human rankings. Addi- tionally, 73% of the images recomposited using our tool are ranked more attractive than their original counterparts by human raters.
Conference Paper
Full-text available
In this paper, we report a study that examines the relationship between image-based computational analyses of web pages and users' aesthetic judgments about the same image material. Web pages were iteratively decomposed into quadrants of minimum entropy (quadtree decomposition) based on low-level image statistics, to permit a characterization of these pages in terms of their respective organizational symmetry, balance and equilibrium. These attributes were then evaluated for their correlation with human participants' subjective ratings of the same web pages on four aesthetic and affective dimensions. Several of these correlations were quite large and revealed interesting patterns in the relationship between low-level (i.e., pixel-level) image statistics and design- relevant dimensions. Author Keywords
Conference Paper
Full-text available
The explosion of image data on the Internet has the potential to foster more sophisticated and robust models and algorithms to index, retrieve, organize and interact with images and multimedia data. But exactly how such data can be harnessed and organized remains a critical problem. We introduce here a new database called ldquoImageNetrdquo, a large-scale ontology of images built upon the backbone of the WordNet structure. ImageNet aims to populate the majority of the 80,000 synsets of WordNet with an average of 500-1000 clean and full resolution images. This will result in tens of millions of annotated images organized by the semantic hierarchy of WordNet. This paper offers a detailed analysis of ImageNet in its current state: 12 subtrees with 5247 synsets and 3.2 million images in total. We show that ImageNet is much larger in scale and diversity and much more accurate than the current image datasets. Constructing such a large-scale database is a challenging task. We describe the data collection scheme with Amazon Mechanical Turk. Lastly, we illustrate the usefulness of ImageNet through three simple applications in object recognition, image classification and automatic object clustering. We hope that the scale, accuracy, diversity and hierarchical structure of ImageNet can offer unparalleled opportunities to researchers in the computer vision community and beyond.
Conference Paper
Full-text available
The visual appearance of a Web page influences the way a user will interact with the page. Web page structural elements (such as text, tables, links, and images) and their characteristics (such as colour and size) are used to determine the visual presentation and complexity level of a Web page. We theorise that by understanding a user's visual and aesthetic perception of a Web page we can understand the cognitive effort required for interaction with that page. This paper describes an investigation into user perception of the visual complexity and aesthetic appearance of Web pages. Results show a strong and high correlation between users' perception of visual complexity, structural elements (links, images, words and sections) and aesthetic appearance (organisation, clearness, cleanliness, interestingness and beautifulness) of a Web page. We argue that the results should be used as a further understanding for keeping the balance between aesthetic appearance of a Web page and its visual complexity. Web pages will then be designed that can still be aesthetically attractive but also usable and not overloaded with information for the users.
Conference Paper
Online peer-to-peer platforms like Airbnb allow hosts to list a property (e.g. a house, or a room) for short-term rentals. In this work, we examine how hosts describe themselves on their Airbnb profile pages. We use a mixed-methods study to develop a categorization of the topics that hosts self-disclose in their profile descriptions, and show that these topics differ depending on the type of guest engagement expected. We also examine the perceived trustworthiness of profiles using topic-coded profiles from 1,200 hosts, showing that longer self-descriptions are perceived to be more trustworthy. Further, we show that there are common strategies (a mix of topics) hosts use in self-disclosure, and that these strategies cause differences in perceived trustworthiness scores. Finally, we show that the perceived trustworthiness score is a significant predictor of host choice--especially for shorter profiles that show more variation. The results are consistent with uncertainty reduction theory, reflect on the assertions of signaling theory, and have important design implications for sharing economy platforms, especially those facilitating online-to-offline social exchange.
Conference Paper
Convolutional networks are at the core of most stateof-the-art computer vision solutions for a wide variety of tasks. Since 2014 very deep convolutional networks started to become mainstream, yielding substantial gains in various benchmarks. Although increased model size and computational cost tend to translate to immediate quality gains for most tasks (as long as enough labeled data is provided for training), computational efficiency and low parameter count are still enabling factors for various use cases such as mobile vision and big-data scenarios. Here we are exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization. We benchmark our methods on the ILSVRC 2012 classification challenge validation set demonstrate substantial gains over the state of the art: 21.2% top-1 and 5.6% top-5 error for single frame evaluation using a network with a computational cost of 5 billion multiply-adds per inference and with using less than 25 million parameters. With an ensemble of 4 models and multi-crop evaluation, we report 3.5% top-5 error and 17.3% top-1 error.
Conference Paper
This paper focuses on a practically very important problem of matching a real-world product photo to exactly the same item(s) in online shopping sites. The task is extremely challenging because the user photos (i.e., the queries in this scenario) are often captured in uncontrolled environments, while the product images in online shops are mostly taken by professionals with clean backgrounds and perfect lighting conditions. To tackle the problem, we study deep network architectures and training schemes, with the goal of learning a robust deep feature representation that is able to bridge the domain gap between the user photos and the online product images. Our contributions are two-fold. First, we propose an alternative of the popular contrastive loss used in siamese deep networks, namely robust contrastive loss, where we "relax" the penalty on positive pairs to alleviate over-fitting. Second, a multi-task fine-tuning approach is introduced to learn a better feature representation, which not only incorporates knowledge from the provided training photo pairs, but also explores additional information from the large ImageNet dataset to regularize the fine-tuning procedure. Experiments on two challenging real-world datasets demonstrate that both the robust contrastive loss and the multi-task fine-tuning approach are effective, leading to very promising results with a time cost suitable for real-time retrieval.
Article
Given two images, we want to predict which exhibits a particular visual attribute more than the other - even when the two images are quite similar. Existing relative attribute methods rely on global ranking functions, yet rarely will the visual cues relevant to a comparison be constant for all data, nor will humans' perception of the attribute necessarily permit a global ordering. To address these issues, we propose a local learning approach for fine-grained visual comparisons. Given a novel pair of images, we learn a local ranking model on the fly, using only analogous training comparisons. We show how to identify these analogous pairs using learned metrics. With results on three challenging datasets - including a large newly curated dataset for fine-grained comparisons - our method outperforms state-of-the-art methods for relative attribute prediction.
Conference Paper
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.
Conference Paper
Users make lasting judgments about a website's appeal within a split second of seeing it for the first time. This first impression is influential enough to later affect their opinions of a site's usability and trustworthiness. In this paper, we demonstrate a means to predict the initial impression of aesthetics based on perceptual models of a website's colorfulness and visual complexity. In an online study, we collected ratings of colorfulness, visual complexity, and visual appeal of a set of 450 websites from 548 volunteers. Based on these data, we developed computational models that accurately measure the perceived visual complexity and colorfulness of website screenshots. In combination with demographic variables such as a user's education level and age, these models explain approximately half of the variance in the ratings of aesthetic appeal given after viewing a website for 500ms only.
With the ever-expanding volume of visual content available, the ability to organize and navigate such content by aesthetic preference is becoming increasingly important. While still in its nascent stage, research into computational models of aesthetic preference already shows great potential. However, to advance research, realistic, diverse and challenging databases are needed. To this end, we introduce a new large-scale database for conducting Aesthetic Visual Analysis: AVA. It contains over 250,000 images along with a rich variety of meta-data including a large number of aesthetic scores for each image, semantic labels for over 60 categories as well as labels related to photographic style. We show the advantages of AVA with respect to existing databases in terms of scale, diversity, and heterogeneity of annotations. We then describe several key insights into aesthetic preference afforded by AVA. Finally, we demonstrate, through three applications, how the large scale of AVA can be leveraged to improve performance on existing preference tasks.
Article
The online merchant of an e-marketplace consists of an intermediary, providing the market infrastructure, and the community of sellers conducting business within that infrastructure. Typically, consumers willingly buy from unknown sellers within an e-marketplace, despite the apparent risk, since they trust the institutional mechanisms furnished by the relatively well-known intermediary. Consumers’ trust in one component of the e-marketplace merchant may not only affect their trust in the other, but also influence the way consumers make online purchases. This paper explores the impact of trust on consumer behavior in e-marketplaces. An empirical study has been conducted to accomplish our research objectives, using a questionnaire survey of 222 active e-marketplace shoppers in Korea. The results reveal that consumer trust in an intermediary has a strong influence upon both attitudinal loyalty and purchase intentions, although consumer trust in the community of sellers has no significant effect on the two constructs representing consumer behavior. In addition, it was found that trust is transferred from an intermediary to the community of sellers, implying that the trustworthiness of the intermediary plays a critical role in determining the extent to which consumers trust and accept the sellers in the e-marketplace. This paper offers some implications from the findings of the research.
Article
This paper relates quality and uncertainty. The existence of goods of many grades poses interesting and important problems for the theory of markets. On the one hand, the interaction of quality differences and uncertainty may explain important institutions of the labor market. On the other hand, this paper presents a struggling attempt to give structure to the statement: “Business in under-developed countries is difficult”; in particular, a structure is given for determining the economic costs of dishonesty. Additional applications of the theory include comments on the structure of money markets, on the notion of “insurability,” on the liquidity of durables, and on brand-name goods.
Conference Paper
Automatic photo quality assessment and selection systems are helpful for managing the large mount of consumer photos. In this paper, we present such a system based on evaluating the aesthetic quality of consumer photos. The proposed system focuses on photos with faces, which constitute an important part of consumer photo albums. The system has three contributions: 1) We propose an aesthetics-based photo assessment algorithm, by considering different aesthetics-related factors, including the technical characteristics of the photo and the specific features related to faces; 2) Based on the aesthetic measurement, we propose a cropping-based photo editing algorithm, which differs from prior works by eliminating unimportant faces before optimizing photo composition; 3) We also incorporate the aesthetic evaluation with other metrics to select quintessential photos for a large collection of photos. The entire system is delivered by a web interface, which allows users to submit images or albums, and returns promising results for photo evaluation, editing recommendation, and photo selection.
Conference Paper
Aesthetics, in the world of art and photography, refers to the principles of the nature and appreciation of beauty. Judging beauty and other aesthetic qualities of photographs is a highly subjective task. Hence, there is no unanimously agreed standard for measuring aesthetic value. In spite of the lack of firm rules, certain features in photographic images are believed, by many, to please humans more than certain others. In this paper, we treat the challenge of automatically inferring aesthetic quality of pictures using their visual content as a machine learning problem, with a peer-rated online photo sharing Website as data source. We extract certain visual features based on the intuition that they can discriminate between aesthetically pleasing and displeasing images. Automated classifiers are built using support vector machines and classification trees. Linear regression on polynomial terms of the features is also applied to infer numerical aesthetics ratings. The work attempts to explore the relationship between emotions which pictures arouse in people, and their low-level content. Potential applications include content-based image retrieval and digital photography.
Conference Paper
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.
Article
Despite the fact that more and more people are selling things online, the community of sellers is under-investigated by information systems researchers. This research explores the role of sellers’ trust in the continued use of online marketplaces. This research differentiates between the sellers’ trust in intermediaries and their trust in the community of buyers. In addition, the concept of trust is examined with a balanced view of cognitive and affective trust. A research model is developed. Empirical data collected from sellers at uBid.com confirm the research model and hypotheses. The findings show that, for online sellers, (1) both cognitive and affective components of trust matter; (2) trust in the intermediary impacts trust in the community of buyers through the trust transference mechanism; (3) trust influences sellers ’ retention to online marketplaces indirectly via perceived usefulness and perceived enjoyment of using online marketplaces; and (4) perceived enjoyment is an important antecedent of sellers’ retention. This research has implications for information systems research and practice.
Article
ABSTRACT For online marketplaces to succeed and prevent a market of lemons, their feedback mechanism (reputation system) must differentiate among sellers and create price premiumsfor trustworthy ones as returns to their reputation. However, the literature has solely focused on numerical (positive and negative) feedback ratings, alas ignoring the role offeedback text comments. These text comments are proposed to convey useful reputation information about a seller’s prior transactions that cannot be fully captured with crude numerical ratings. Building upon the economics and trust literatures, this study examines the rich content of feedback text comments and their role in building buyer’s trust in a seller’s benevolence and credibility. In turn, benevolence and credibility are proposed to differentiate among sellers by influencing the price premiums that a seller receives from buyers. This paper utilizes content analysis to quantify over 10,000 publicly-available feedback text comments of 420 sellers in eBay’s online auction marketplace, and match them withprimary data from 420 buyers that recently transacted with these 420 sellers. These dyadic data show that evidence of extraordinarypast seller behavior contained in the sellers’ feedback text comments,createsprice premiums for reputable sellers by engendering buyer’s trust in the sellers’ benevolence and credibility (controlling for the impact of numerical ratings). The addition of text comments and benevolence helps explain a greater variance in price premiums (R,=20-30%). By showing the economic value of feedback text comments through trust in a seller’s benevolence and credibility, this study helps explain the success of online marketplaces thatprimarily rely on the text comments (versus crude numerical ratings) to differentiateamong,sellers and prevent a market of ‘lemon’ sellers.
Article
Increasing use of the World Wide Web as a B2C commercial tool raises interest in understanding the key issues in building relationships with customers on the Internet. Trust is believed to be the key to these relationships. Given the differences between a virtual and a conventional marketplace, antecedents and consequences of trust merit re-examination. This research identifies a number of key factors related to trust in the B2C context and proposes a framework based on a series of underpinning relationships among these factors. The findings in this research suggest that people are more likely to purchase from the web if they perceive a higher degree of trust in e-commerce and have more experience in using the web. Customer’s trust levels are likely to be influenced by the level of perceived market orientation, site quality, technical trustworthiness, and user’s web experience. People with a higher level of perceived site quality seem to have a higher level of perceived market orientation and trustworthiness towards e-commerce. Furthermore, people with a higher level of trust in e-commerce are more likely to participate in e-commerce. Positive ‘word of mouth’, money back warranty and partnerships with well-known business partners, rank as the top three effective risk reduction tactics. These findings complement the previous findings on e-commerce and shed light on how to establish a trust relationship on the World Wide Web.
The impact of images on user clicks in product search
  • S H Chung
  • A Goswami
  • H Lee
  • J Hu
S. H. Chung, A. Goswami, H. Lee, and J. Hu. The impact of images on user clicks in product search. In Proceedings of the 12th International Workshop on Multimedia Data Mining, pages 25-33. ACM, 2012.
Rapid: Rating pictorial aesthetics using deep learning
  • X Lu
  • Z Lin
  • H Jin
  • J Yang
  • J Z Wang
X. Lu, Z. Lin, H. Jin, J. Yang, and J. Z. Wang. Rapid: Rating pictorial aesthetics using deep learning. In Proceedings of the 22nd ACM International Conference on Multimedia, pages 457-466. ACM, 2014.
Automatic differentiation in pytorch
  • A Paszke
  • S Gross
  • S Chintala
  • G Chanan
  • E Yang
  • Z De-Vito
  • Z Lin
  • A Desmaison
  • L Antiga
  • A Lerer
A. Paszke, S. Gross, S. Chintala, G. Chanan, E. Yang, Z. De-Vito, Z. Lin, A. Desmaison, L. Antiga, and A. Lerer. Automatic differentiation in pytorch. In Proceedings of the Workshop on Neural Information Processing Systems, 2017.
Grabcut: Interactive foreground extraction using iterated graph cuts
  • C Rother
  • V Kolmogorov
  • A Blake
C. Rother, V. Kolmogorov, and A. Blake. Grabcut: Interactive foreground extraction using iterated graph cuts. In Proceedings of the ACM Transactions on Graphics, volume 23, pages 309-314. ACM, 2004.