共找到 20 条结果
Signals are short textual or visual snippets displayed on the eBay View-Item (VI) page, providing additional, contextual information for users about the viewed item. The aim in displaying the signals is to facilitate intelligent purchase and to incentivise engagement. In this paper, we present two approaches for developing statistical models that optimally populate the VI page with signals. Both approaches were A/B tested, and yielded significant increase in business metrics.
E-commerce sellers are recommended keyphrases based on their inventory on which they advertise to increase buyer engagement (clicks/sales). The relevance of advertiser keyphrases plays an important role in preventing the inundation of search systems with numerous irrelevant items that compete for attention in auctions, in addition to maintaining a healthy seller perception. In this work, we describe the shortcomings of training Advertiser keyphrase relevance filter models on click/sales/search relevance signals and the importance of aligning with human judgment, as sellers have the power to adopt or reject said keyphrase recommendations. In this study, we frame Advertiser keyphrase relevance as a complex interaction between 3 dynamical systems -- seller judgment, which influences seller adoption of our product, Advertising, which provides the keyphrases to bid on, and Search, who holds the auctions for the same keyphrases. This study discusses the practicalities of using human judgment via a case study at eBay Advertising and demonstrate that using LLM-as-a-judge en-masse as a scalable proxy for seller judgment to train our relevance models achieves a better harmony across the thre
Detecting counterfeits on online marketplaces is challenging, and current methods struggle with the volume of sales on platforms like eBay, while cryptomarkets openly sell counterfeits. Leveraging information from 453 cryptomarket counterfeits, we automated a search for corresponding products on eBay, utilizing image and text similarity metrics. We collected data twice over 4-months to analyze changes with an average of 159 eBay products per cryptomarket item, totaling 134k products. We found identical products, which would warrant further investigation as to whether they are counterfeits. Results indicate increasing difficulty finding similar products over time, moderated by product type and origin. Future improved versions of the current system could be used to examine possible connections between cryptomarket and surface web listings more closely and could hold practical value in supporting the detection of counterfeits on the surface web.
We introduce the LiLiuM series of large language models (LLMs): 1B, 7B, and 13B parameter models developed 100% in-house to fit eBay's specific needs in the e-commerce domain. This gives eBay full control over all aspects of the models including license, data, vocabulary, and architecture. We expect these models to be used as a foundation for fine-tuning and instruction-tuning, eliminating dependencies to external models. The LiLiuM LLMs have been trained on 3 trillion tokens of multilingual text from general and e-commerce domain. They perform similar to the popular LLaMA-2 models on English natural language understanding (NLU) benchmarks. At the same time, we outperform LLaMA-2 on non-English NLU tasks, machine translation and on e-commerce specific downstream tasks. As part of our data mixture, we utilize the newly released RedPajama-V2 dataset for training and share our insights regarding data filtering and deduplication. We also discuss in detail how to serialize structured data for use in autoregressive language modeling. We provide insights on the effects of including code and parallel machine translation data in pre-training. Furthermore, we develop our own tokenizer and mo
E-commerce sellers are advised to bid on keyphrases to boost their advertising campaigns. These keyphrases must be relevant to prevent irrelevant items from cluttering Search systems and to maintain positive seller perception. It is vital that keyphrase suggestions align with seller, Search, and buyer judgments. Given the challenges in collecting negative feedback in these systems, LLMs have been used as a scalable proxy for human judgments. We present an empirical study on a major e-commerce platform of a distillation framework involving an LLM teacher, a cross-encoder assistant and a bi-encoder Embedding Based Retrieval (EBR) student model, aimed at mitigating click-induced biases and provide more diverse keyphrase recommendations while aligning advertising, search and buyer preferences.
In this paper, we propose a novel end-to-end approach for scalable visual search infrastructure. We discuss the challenges we faced for a massive volatile inventory like at eBay and present our solution to overcome those. We harness the availability of large image collection of eBay listings and state-of-the-art deep learning techniques to perform visual search at scale. Supervised approach for optimized search limited to top predicted categories and also for compact binary signature are key to scale up without compromising accuracy and precision. Both use a common deep neural network requiring only a single forward inference. The system architecture is presented with in-depth discussions of its basic components and optimizations for a trade-off between search relevance and latency. This solution is currently deployed in a distributed cloud infrastructure and fuels visual search in eBay ShopBot and Close5. We show benchmark on ImageNet dataset on which our approach is faster and more accurate than several unsupervised baselines. We share our learnings with the hope that visual search becomes a first class citizen for all large scale search engines rather than an afterthought.
At eBay, there are thousands of product health metrics for different domain teams to monitor. We built a two-phase alerting system to notify users with actionable alerts based on anomaly detection and alert retrieval. In the first phase, we developed an efficient anomaly detection algorithm, called Moving Metric Detector (MMD), to identify potential alerts among metrics with distribution agnostic criteria. In the second alert retrieval phase, we built additional logic with feedbacks to select valid actionable alerts with point-wise ranking model and business rules. Compared with other trend and seasonality decomposition methods, our decomposer is faster and better to detect anomalies in unsupervised cases. Our two-phase approach dramatically improves alert precision and avoids alert spamming in eBay production.
eBay sent bloody pig mask and other macabre deliveries to journalists' home
For months on end, eBay employees and contractors made life hell for a couple that had criticized the company。 Six years later, the company is finally paying up
Market segmentation of an online auction site is studied by analyzing the users' bidding behavior. The distribution of user activity is investigated and a network of bidders connected by common interest in individual articles is constructed. The network's cluster structure corresponds to the main user groups according to common interest, exhibiting hierarchy and overlap. Key feature of the analysis is its independence of any similarity measure between the articles offered on eBay, as such a measure would only introduce bias in the analysis. Results are compared to null models based on random networks and clusters are validated and interpreted using the taxonomic classifications of eBay categories. We find clear-cut and coherent interest profiles for the bidders in each cluster. The interest profiles of bidder groups are compared to the classification of articles actually bought by these users during the time span 6-9 months after the initial grouping. The interest profiles discovered remain stable, indicating typical interest profiles in society. Our results show how network theory can be applied successfully to problems of market segmentation and sociological milieu studies with s
Online auctions are fast gaining popularity in today's electronic commerce. Relative to offline auctions, there is a greater degree of multiple bidding and late bidding in online auctions, an empirical finding by some recent research. These two behaviors (multiple bidding and late bidding) are of ``strategic'' importance to online auctions and hence important to investigate. In this article we empirically measure the distribution of bid timings and the extent of multiple bidding in a large set of online auctions, using bidder experience as a mediating variable. We use data from the popular auction site \url{www.eBay.com} to investigate more than 10,000 auctions from 15 consumer product categories. We estimate the distribution of late bidding and multiple bidding, which allows us to place these product categories along a continuum of these metrics (the extent of late bidding and the extent of multiple bidding). Interestingly, the results of the analysis distinguish most of the product categories from one another with respect to these metrics, implying that product categories, after controlling for bidder experience, differ in the extent of multiple bidding and late bidding observed
Recommender systems are an essential component of e-commerce marketplaces, helping consumers navigate massive amounts of inventory and find what they need or love. In this paper, we present an approach for generating personalized item recommendations in an e-commerce marketplace by learning to embed items and users in the same vector space. In order to alleviate the considerable cold-start problem present in large marketplaces, item and user embeddings are computed using content features and multi-modal onsite user activity respectively. Data ablation is incorporated into the offline model training process to improve the robustness of the production system. In offline evaluation using a dataset collected from eBay traffic, our approach was able to improve the Recall@k metric over the Recently-Viewed-Item (RVI) method. This approach to generating personalized recommendations has been launched to serve production traffic, and the corresponding scalable engineering architecture is also presented. Initial A/B test results show that compared to the current personalized recommendation module in production, the proposed method increases the surface rate by $\sim$6\% to generate recommenda
eBay partners with external vendors, which allows customers to freely select a vendor to complete their eBay experiences. However, vendor outages can hinder customer experiences. Consequently, eBay can disable a problematic vendor to prevent customer loss. Disabling the vendor too late risks losing customers willing to switch to other vendors, while disabling it too early risks losing those unwilling to switch. In this paper, we propose a data-driven solution to answer whether eBay should disable a problematic vendor and when to disable it. Our solution involves forecasting customer behavior. First, we use a multiplicative seasonality model to represent behavior if all vendors are fully functioning. Next, we use a Monte Carlo simulation to represent behavior if the problematic vendor remains enabled. Finally, we use a linear model to represent behavior if the vendor is disabled. By comparing these forecasts, we determine the optimal time for eBay to disable the problematic vendor.
In the domain of sponsored search advertising, the focus of Keyphrase recommendation has largely been on exact match types, which pose issues such as high management expenses, limited targeting scope, and evolving search query patterns. Alternatives like Broad match types can alleviate certain drawbacks of exact matches but present challenges like poor targeting accuracy and minimal supervisory signals owing to limited advertiser usage. This research defines the criteria for an ideal broad match, emphasizing on both efficiency and effectiveness, ensuring that a significant portion of matched queries are relevant. We propose BroadGen, an innovative framework that recommends efficient and effective broad match keyphrases by utilizing historical search query data. Additionally, we demonstrate that BroadGen, through token correspondence modeling, maintains better query stability over time. BroadGen's capabilities allow it to serve daily, millions of sellers at eBay with over 2.5 billion items.
E-commerce sellers are recommended keyphrases based on their inventory on which they advertise to increase buyer engagement (clicks/sales). Keyphrases must be pertinent to items; otherwise, it can result in seller dissatisfaction and poor targeting -- towards that end relevance filters are employed. In this work, we describe the shortcomings of training relevance filter models on biased click/sales signals. We re-conceptualize advertiser keyphrase relevance as interaction between two dynamical systems -- Advertising which produces the keyphrases and Search which acts as a middleman to reach buyers. We discuss the bias of search relevance systems (middleman bias) and the need to align advertiser keyphrases with search relevance signals. We also compare the performance of cross encoders and bi-encoders in modeling this alignment and the scalability of such a solution for sellers at eBay.
There are unique challenges to developing item recommender systems for e-commerce platforms like eBay due to sparse data and diverse user interests. While rich user-item interactions are important, eBay's data sparsity exceeds other e-commerce sites by an order of magnitude. To address this challenge, we propose CoActionGraphRec (CAGR), a text based two-tower deep learning model (Item Tower and User Tower) utilizing co-action graph layers. In order to enhance user and item representations, a graph-based solution tailored to eBay's environment is utilized. For the Item Tower, we represent each item using its co-action items to capture collaborative signals in a co-action graph that is fully leveraged by the graph neural network component. For the User Tower, we build a fully connected graph of each user's behavior sequence, with edges encoding pairwise relationships. Furthermore, an explicit interaction module learns representations capturing behavior interactions. Extensive offline and online A/B test experiments demonstrate the effectiveness of our proposed approach and results show improved performance over state-of-the-art methods on key metrics.
Static and hard-coded layer-two network identifiers are well known to present security vulnerabilities and endanger user privacy. In this work, we introduce a new privacy attack against Wi-Fi access points listed on secondhand marketplaces. Specifically, we demonstrate the ability to remotely gather a large quantity of layer-two Wi-Fi identifiers by programmatically querying the eBay marketplace and applying state-of-the-art computer vision techniques to extract IEEE 802.11 BSSIDs from the seller's posted images of the hardware. By leveraging data from a global Wi-Fi Positioning System (WPS) that geolocates BSSIDs, we obtain the physical locations of these devices both pre- and post-sale. In addition to validating the degree to which a seller's location matches the location of the device, we examine cases of device movement -- once the device is sold and then subsequently re-used in a new environment. Our work highlights a previously unrecognized privacy vulnerability and suggests, yet again, the strong need to protect layer-two network identifiers.
Electronic commerce, or e-commerce, is the buying and selling of goods and services, or the transmitting of funds or data online. E-commerce platforms come in many kinds, with global players such as Amazon, Airbnb, Alibaba, eBay and platforms targeting specific geographic regions. Information retrieval has a natural role to play in e-commerce, especially in connecting people to goods and services. Information discovery in e-commerce concerns different types of search (e.g., exploratory search vs. lookup tasks), recommender systems, and natural language processing in e-commerce portals. The rise in popularity of e-commerce sites has made research on information discovery in e-commerce an increasingly active research area. This is witnessed by an increase in publications and dedicated workshops in this space. Methods for information discovery in e-commerce largely focus on improving the effectiveness of e-commerce search and recommender systems, on enriching and using knowledge graphs to support e-commerce, and on developing innovative question answering and bot-based solutions that help to connect people to goods and services. In this survey, an overview is given of the fundamental
Consumer demand forecasting is of high importance for many e-commerce applications, including supply chain optimization, advertisement placement, and delivery speed optimization. However, reliable time series sales forecasting for e-commerce is difficult, especially during periods with many anomalies, as can often happen during pandemics, abnormal weather, or sports events. Although many time series algorithms have been applied to the task, prediction during anomalies still remains a challenge. In this work, we hypothesize that leveraging external knowledge found in world events can help overcome the challenge of prediction under anomalies. We mine a large repository of 40 years of world events and their textual representations. Further, we present a novel methodology based on transformers to construct an embedding of a day based on the relations of the day's events. Those embeddings are then used to forecast future consumer behavior. We empirically evaluate the methods over a large e-commerce products sales dataset, extracted from eBay, one of the world's largest online marketplaces. We show over numerous categories that our method outperforms state-of-the-art baselines during ano
Online sellers and advertisers are recommended keyphrases for their listed products, which they bid on to enhance their sales. One popular paradigm that generates such recommendations is Extreme Multi-Label Classification (XMC), which involves tagging/mapping keyphrases to items. We outline the limitations of using traditional item-query based tagging or mapping techniques for keyphrase recommendations on E-Commerce platforms. We introduce GraphEx, an innovative graph-based approach that recommends keyphrases to sellers using extraction of token permutations from item titles. Additionally, we demonstrate that relying on traditional metrics such as precision/recall can be misleading in practical applications, thereby necessitating a combination of metrics to evaluate performance in real-world scenarios. These metrics are designed to assess the relevance of keyphrases to items and the potential for buyer outreach. GraphEx outperforms production models at eBay, achieving the objectives mentioned above. It supports near real-time inferencing in resource-constrained production environments and scales effectively for billions of items.