User Click Research Articles

Behind the clicks: Can Amazon allocate user attention as it pleases?

We investigate Amazon's ability to direct user clicks to more visually prominent search results, even as quality declines with the increasing prevalence of sponsored advertising products. We analyze product results from over 2,000 Amazon Marketplace search queries to estimate how a top three most clicked product's features (price and quality) and screen placement influence a user's clicks. Our econometric results show that the position of a product search result (“position bias”), adjusted for its relative prominence on the screen, strongly shapes whether a user clicks on it. Within the top five search results, where typically four are advertisements, users exhibit decreased sensitivity to a product's relevance or pricing. This allows Amazon's sponsored ads to leverage product prominence as a mechanism for rent extraction from users and producers. Regulatory frameworks might limit platforms from exploiting consumers' satisficing behaviour online, including via moderating excessive advertising in algorithmic search results.

Heatmaps: A Tool for Studying Book Selection Preferences on Discovery Platforms

ABSTRACTThis study evaluates the usefulness of heatmaps in understanding user preferences during book selection in online discovery platforms. It compares two heatmaps generated from user clicks and eye‐tracking data to identify key metadata influencing readers' choices. Initial results indicate that click heatmaps can be a useful tool for this type of investigation, especially considering the cost of eye‐tracking, which is typically used to generate attention heatmaps.

AeroClick: An advanced single-click interactive framework for aeroengine defect segmentation

In aeroengine defect inspection, traditional algorithms are often encumbered by overreliance on pixel-level labels and an inability to adapt to variations in domain-specific data, resulting in a lack of efficiency and generalizability. To address these challenges, this paper proposes an innovative solution using a single-click interactive framework that combines user-friendly interaction and computational efficiency. First, user clicks are incorporated into an additional channel coupled with the original images. Second, superpixel-guided interactive signals and corresponding Gaussian heat maps are utilized to embed user clicks into the training and prediction processes, tolerating certain click errors without additional corrections. Third, the tailored backbone handles defect variations and captures intricate details. Furthermore, the Bayesian-optimization-based refinement process further enhances accuracy and generalizability. The experimental results from four industrial scenarios demonstrate that our method not only excels in aeroengine defect segmentation but also has the potential to be effectively applied as a semi-automatic annotation tool in other industrial inspection fields.

Evaluating User Engagement in Online News: A Deep Learning Approach Based on Attractiveness and Multiple Features

Online news platforms have become users’ primary information sources. However, they focus on attracting users to click on the news and ignore whether the news triggers a sense of engagement, which could potentially reduce users’ participation in public events. Therefore, this study constructs four indicators by assessing user engagement to build an intelligent system to help platforms optimize their publishing strategies. First, this study defines user engagement evaluation as a classification task that divides user engagement into four indicators and proposes an extended LDA model based on user click–comment behavior (UCCB), using which the attractiveness of words in news headlines and content can be effectively represented. Second, this study proposes a deep user engagement evaluation (DUEE) model that integrates news attractiveness and multiple features in an attention-based deep neural network for user engagement evaluation. The DUEE model considers various elements that collectively determine the ability of the news to attract clicks and engagement. Third, the proposed model is compared with the baseline and state-of-the-art techniques, showing that it outperforms all existing methods. This study provides new research contributions and ideas for improving user engagement in online news evaluation.

ReCRec: Reasoning the Causes of Implicit Feedback for Debiased Recommendation

Implicit feedback ( e.g ., user clicks) is widely used in building recommender systems (RS). However, the inherent notorious exposure bias significantly affects recommendation performance. Exposure bias refers a phenomenon that implicit feedback is influenced by user exposure, and does not precisely reflect user preference. Current methods for addressing exposure bias primarily reduce confidence in unclicked data, employ exposure models, or leverage propensity scores. Regrettably, these approaches often lead to biased estimations or elevated model variance, yielding sub-optimal results. To overcome these limitations, we propose a new method ReCRec that Re asons the C auses behind the implicit feedback for debiased Rec ommendation. ReCRec identifies three scenarios behind unclicked data — i.e. , unexposed, dislike or a combination of both. A reasoning module is employed to infer the category to which each instance pertains. Consequently, the model is capable of extracting reliable positive and negative signals from unclicked data, thereby facilitating more accurate learning of user preferences. We also conduct thorough theoretical analyses to demonstrate the debiased nature and low variance of ReCRec. Extensive experiments on both semi-synthetic and real-world datasets validate its superiority over state-of-the-art methods.

Age-Gating and Marketing Differences Between Storefront and Non-Storefront Cannabis Retailers

Objective: The study investigated whether California storefront and non-storefront cannabis retailers are adhering to online age-gating requirements and whether differences in website marketing practices exist. Methods: Websites of 134 storefront and 115 non-storefront licensed retailers were randomly selected. Bivariate associations were tested between retailer type and website marketing, age-gating methods, and presence of age-gating at various purchase stages. Results: Among the 200 (80.3%) websites with age-gating when entering, 182 (91%) employed an ineffective method where users click either “Yes” or “No” to confirm their age. Moreover, 49 (19.68%) websites lacked age-gating when entering. Amongst those requiring photo identification during checkout (n = 100, 40.16%), 97% allowed users to proceed after uploading an irrelevant image. Significantly more storefront retailers employed combined age-gating at entry, mandatory account registration, and age-gating during checkout than non-storefront retailers (X2 (1, N = 249) = 7.69, p < .01). Retailer websites frequently displayed “clean” labels (n = 200, 80.32%), followed by positive state claims (n = 198, 79.52%), physical health claims (n = 166, 66.67%), and mental health claims (n = 146, 58.63%). Significantly more storefront retailers displayed physical health claims (X2 (1, N = 249) = 7.52, p < .01) and health warnings than non-storefront retailers (X2 (1, N = 249) = 4.13, p = .04). Conclusions: Most cannabis retailers comply with age-gating requirements; however, methods employed are easily circumvented. Youths’ easy and unrestricted access to cannabis retailer websites may increase positive attitudes about cannabis and encourage use.

TETRIS: Towards Exploring the Robustness of Interactive Segmentation

Interactive segmentation methods rely on user inputs to iteratively update the selection mask. A click specifying the object of interest is arguably the most simple and intuitive interaction type, and thereby the most common choice for interactive segmentation. However, user clicking patterns in the interactive segmentation context remain unexplored. Accordingly, interactive segmentation evaluation strategies rely more on intuition and common sense rather than empirical studies (e.g., assuming that users tend to click in the center of the area with the largest error). In this work, we conduct a real-user study to investigate real user clicking patterns. This study reveals that the intuitive assumption made in the common evaluation strategy may not hold. As a result, interactive segmentation models may show high scores in the standard benchmarks, but it does not imply that they would perform well in a real world scenario. To assess the applicability of interactive segmentation methods, we propose a novel evaluation strategy providing a more comprehensive analysis of a model's performance. To this end, we propose a methodology for finding extreme user inputs by a direct optimization in a white-box adversarial attack on the interactive segmentation model. Based on the performance with such adversarial user inputs, we assess the robustness of interactive segmentation models w.r.t click positions. Besides, we introduce a novel benchmark for measuring the robustness of interactive segmentation, and report the results of an extensive evaluation of dozens of models.

CFR-ICL: Cascade-Forward Refinement with Iterative Click Loss for Interactive Image Segmentation

The click-based interactive segmentation aims to extract the object of interest from an image with the guidance of user clicks. Recent work has achieved great overall performance by employing feedback from the output. However, in most state-of-the-art approaches, 1) the inference stage involves inflexible heuristic rules and requires a separate refinement model, and 2) the number of user clicks and model performance cannot be balanced. To address the challenges, we propose a click-based and mask-guided interactive image segmentation framework containing three novel components: Cascade-Forward Refinement (CFR), Iterative Click Loss (ICL), and SUEM image augmentation. The CFR offers a unified inference framework to generate segmentation results in a coarse-to-fine manner. The proposed ICL allows model training to improve segmentation and reduce user interactions simultaneously. The proposed SUEM augmentation is a comprehensive way to create large and diverse training sets for interactive image segmentation. Extensive experiments demonstrate the state-of-the-art performance of the proposed approach on five public datasets. Remarkably, our model reduces by 33.2%, and 15.5% the number of clicks required to surpass an IoU of 0.95 in the previous state-of-the-art approach on the Berkeley and DAVIS sets, respectively.

TMH: Two-Tower Multi-Head Attention neural network for CTR prediction.

Click-through rate (CTR) prediction is a term used to predict the probability of a user clicking on an ad or item and has become a popular research area in advertising. As the volume of Internet data increases, the labor costs of traditional feature engineering continue to rise. To reduce the dependence on feature interactions, this paper proposes a fusion model that combines explicit and implicit feature interactions, called the Two-Tower Multi-Head Attention Neural Network (TMH) approach. The model integrates multiple components such as multi-head attention, residual network, and deep neural networks into an end-to-end model that automatically obtains vector-level combinations of explicit and implicit features to predict click-through rates through higher-order explicit and implicit interactions. We evaluated the effectiveness of TMH in CTR prediction through numerous experiments using three real datasets. The results demonstrate that our proposed method not only outperforms existing prediction methods but also offers good interpretability.

A Hotspot-Aware Personalized News Recommendation Mechanism Based on DistilBERT-TC-MA

Aiming at the problems of existing news recommendation methods, such as inadequate exploration of the semantic information of news, neglecting potential hotspot features of news, and challenging the balance between user preferences and hotspot features, a hotspot-aware personalized news recommendation model (DistilBERT-TC-MA) is suggested, which integrates the distilled version of BERT (DistilBERT), text convolutional neural network (TextCNN), and multilayer attention (MA). First, it takes full advantage of DistilBERT, TextCNN, and self-attention mechanism to achieve news encoding. Following this, representations of trending news are dynamically aggregated using the attention mechanism, while user preferences are mined utilizing user click history. Finally, in order to successfully accomplish the click prediction of candidate news, the hotspot features, user preferences, and candidate news are ultimately combined using a click predictor. The experimental results of the suggested DistilBERT-TC-MA model on MIND dataset are better than several other advanced methods.

Data-driven smoothing approaches for interest modeling in recommendation systems

In recommendation systems, users often click on some items that are distinct from historically clicked items. This verifies the existence of interest gaps between the historical interests reflected by historical behaviors and current interests. The gaps pose key challenges for interest modeling in recommendation systems, i.e., learn user current interests from historical behaviors. To bridge interest gaps, various solutions have been proposed from a model perspective, and mainly focus on capturing interest diversity and interest evolution process (the changes in user interests over time). Different from existing solutions, our key idea is smoothing the historical interests from a data perspective, so that interest gaps can be smoothed and bridged. To realize this idea, we propose two types of data-driven smoothing approaches based on the principle of collaborative filtering, i.e., click-centric smoothing and user-centric smoothing. The approaches can be easily applied to a number of existing solutions and enhance their performances. To ensure that the smoothed interests completely and accurately cover user current interests (coverage and accuracy), the click-centric smoothing leverages the relevant clicks of a click to smooth the interests on the click, and the user-centric smoothing leverages the relevant clicks of a user click sequence to smooth the interests learned from the click sequence. The “smooth” is to enhance the coverage, and “relevant” is to enhance the accuracy. To search the relevant clicks, we develop an iterative Bayesian inference over a click-flow graph, i.e., a new implementation of collaborative filtering. To further enhance the accuracy, we propose new constrained attention networks to generate high-quality smoothing factors from these relevant clicks. Both offline and online experiments show the data-driven smoothing approaches significantly outperform the baselines. The user-centric smoothing has been deployed on the e-commerce platform of Meituan and achieved 1.51% elevation of Click-through & Conversion Rate and 0.53% elevation of Click-through Rate.

Deep Image Matting With Sparse User Interactions.

Image matting is a fundamental and challenging problem in computer vision and graphics. Most existing matting methods leverage a user-supplied trimap as an auxiliary input to produce good alpha matte. However, obtaining high-quality trimap itself is arduous. Recently, some hint-free methods have emerged, however, the matting quality is still far behind the trimap-based methods. The main reason is that, some hints for removing semantic ambiguity and improving matting quality are essential. Apparently, there is a trade-off between interaction cost and matting quality. To balance performance and user-friendliness, we propose an improved deep image matting framework which is trimap-free and only needs sparse user click or scribble interaction to minimize the needed auxiliary constraints while still allowing interactivity. Moreover, we introduce uncertainty estimation that predicts which parts need polishing and conduct uncertainty-guided refinement. To trade off runtime against refinement quality, users can also choose different refinement modes. Experimental results show that our method performs better than existing trimap-free methods and comparably to state-of-the-art trimap-based methods with minimal user effort. Finally, we demonstrate the extensibility of our framework to video human matting without any structure modification, by adding optical flow-based sparse hint propagation and temporal consistency regularization imposed on the single frame.

Efficient click-based interactive segmentation for medical image with improved Plain-ViT.

The primary objective of interactive medical image segmentation systems is to achieve more precise segmentation outcomes with reduced human intervention. This endeavor holds significant clinical importance for both pre-diagnostic pathological assessments and prognostic recovery. Among the various interaction methods available, click-based interactions stand out as an intuitive and straightforward approach compared to alternatives such as graffiti, bounding boxes, and extreme points. To improve the model's ability to interpret click-based interactions, we propose a comprehensive interactive segmentation framework that leverages an iterative weighted loss function based on user clicks. To enhance the segmentation capabilities of the Plain-ViT backbone, we introduce a Residual Multi-Headed Self-Attention encoder with hierarchical inputs and residual connections, offering multiple perspectives on the data. This innovative architecture leads to a remarkable improvement in segmentation model performance. In this research paper, we assess the robustness of our proposed framework using a self-compiled T2-MRI image dataset of the prostate and three publicly available datasets containing images of other organs. Our experimental results convincingly demonstrate that our segmentation model surpasses existing state-of-the-art methods. Furthermore, the incorporation of an iterative loss function training strategy significantly accelerates the model's convergence rate during interactions. In the prostate dataset, we achieved an impressive Intersection over Union (IoU) score of 88.11% and Number of Clicks(NoC) at 80% are 7.03 clicks.

Mining Semantic Correlations Between Mispredictions and Corrections for Interactive Semantic Segmentation.

Interactive semantic segmentation pursues high-quality segmentation results at the cost of a small number of user clicks. It is attracting more and more research attention for its convenience in labeling semantic pixel-level data. Existing interactive segmentation methods often pursue higher interaction efficiency by mining the latent information of user clicks or exploring efficient interaction manners. However, these works neglect to explicitly exploit the semantic correlations between user corrections and model mispredictions, thus suffering from two flaws. First, similar prediction errors frequently occur in actual use, causing users to repeatedly correct them. Second, the interaction difficulty of different semantic classes varies across images, but existing models use monotonic parameters for all images which lack semantic pertinence. Therefore, in this article, we explore the semantic correlations existing in corrections and mispredictions by proposing a simple yet effective online learning solution to the above problems, named correction-misprediction correlation mining ( CM2 ). Specifically, we leverage the correction-misprediction similarities to design a confusion memory module (CMM) for automatic correction when similar prediction errors reappear. Furthermore, we measure the semantic interaction difficulty by counting the correction-misprediction pairs and design a challenge adaptive convolutional layer (CACL), which can adaptively switch different parameters according to interaction difficulties to better segment the challenging classes. Our method requires no extra training besides the online learning process and can effectively improve interaction efficiency. Our proposed CM2 achieves state-of-the-art results on three public semantic segmentation benchmarks.

MedCPT: Contrastive Pre-trained Transformers with large-scale PubMed search logs for zero-shot biomedical information retrieval.

Information retrieval (IR) is essential in biomedical knowledge acquisition and clinical decision support. While recent progress has shown that language model encoders perform better semantic retrieval, training such models requires abundant query-article annotations that are difficult to obtain in biomedicine. As a result, most biomedical IR systems only conduct lexical matching. In response, we introduce MedCPT, a first-of-its-kind Contrastively Pre-trained Transformer model for zero-shot semantic IR in biomedicine. To train MedCPT, we collected an unprecedented scale of 255 million user click logs from PubMed. With such data, we use contrastive learning to train a pair of closely integrated retriever and re-ranker. Experimental results show that MedCPT sets new state-of-the-art performance on six biomedical IR tasks, outperforming various baselines including much larger models, such as GPT-3-sized cpt-text-XL. In addition, MedCPT also generates better biomedical article and sentence representations for semantic evaluations. As such, MedCPT can be readily applied to various real-world biomedical IR tasks. The MedCPT code and model are available at https://github.com/ncbi/MedCPT.

NLP BASED TEXT SUMMARIZATION USING BART MODEL

User interest and response prediction are critical tasks in online advertising, Advertisers may more effectively target their advertising, increase their click-through rates (CTRs), and boost their conversion rates by accurately predicting user interests and behaviors. In recent years, deep learning has developed into a potent tool for predicting consumer interest and response. Deep learning models can accurately predict user interests and responses because they are able to understand complicated patterns from huge quantities of user behavior data. In this study, we suggest a deep learning-based system for online advertising response and interest prediction. Two deep-learning models make up our framework: Thismodel forecasts the likelihood that a user will click on an advertisement. Model for predicting user interest: This model forecasts the likelihood that a user would click on a particular kind of advertising campaign. The models are developed using data from a sizable datasetof user behavior, such as user clicks, ad views, anddemographic data. Compared to conventional approaches, our system has several advantages for predicting user interest and response. First, unlike existing methods, our system can identify Key Words: Deep learning, LST model, user interest prediction, response prediction.

SimpleClick: Interactive Image Segmentation with Simple Vision Transformers.

Click-based interactive image segmentation aims at extracting objects with a limited user clicking. A hierarchical backbone is the de-facto architecture for current methods. Recently, the plain, non-hierarchical Vision Transformer (ViT) has emerged as a competitive backbone for dense prediction tasks. This design allows the original ViT to be a foundation model that can be finetuned for downstream tasks without redesigning a hierarchical backbone for pretraining. Although this design is simple and has been proven effective, it has not yet been explored for interactive image segmentation. To fill this gap, we propose SimpleClick, the first interactive segmentation method that leverages a plain backbone. Based on the plain backbone, we introduce a symmetric patch embedding layer that encodes clicks into the backbone with minor modifications to the backbone itself. With the plain backbone pretrained as a masked autoencoder (MAE), SimpleClick achieves state-of-the-art performance. Remarkably, our method achieves 4.15 NoC@90 on SBD, improving 21.8% over the previous best result. Extensive evaluation on medical images demonstrates the generalizability of our method. We provide a detailed computational analysis, highlighting the suitability of our method as a practical annotation tool.

A Study of Real-Time Advertising Bidding Based on Reinforcement Learning

The purpose of real-time bidding (RTB), which has grown to be a significant component of Internet advertising, is to properly forecast the likelihood of a user clicking on an ad and choose the best bid for that impression. Reinforcement learning (RL) has emerged as a promising approach for RTB bidding strategies due to its ability to learn from feedback and optimize performance over time. This paper surveys RL-based real-time advertising bidding, introduces several main reinforcement learning-based real-time advertising bidding strategies, and explains the advantages of each of these strategies. Furthermore, the author analyzes the current trends and which strategies are combined and why. Overall, this survey sheds light on the potential of RL-based bidding to enhance the effectiveness and efficiency of RTB advertising while also offering information about the current state of the field and future research directions.

Fraud Detection of AD Clicks Using Machine Learning Techniques

Although all businesses face the possibility of fraud, those that rely on internet advertising face an especially high risk of click fraud, which may lead to inaccurate click statistics and unnecessary expenditures. The cost per click for advertising channels might skyrocket if enough people click on the ads. Internet advertising is becoming a significant revenue source for many websites. Under this model, advertisers pay the publisher a flat rate for each click-through from the ad to the advertiser's site. Since spending much on Internet advertising requires significant resources, the term "click fraud" refers to an attack tactic in which the perpetrator repeatedly clicks on a single link for the sole purpose of generating illicit revenue. By clicking on a pay-per-click (PPC) ad many times using a script, fraudsters may trick online advertisers into paying for clicks that never happened. We may use a variety of methods to identify click fraud anytime a human or computer program clicks on a particular link, and then use the click-through rate to ascertain whether the clicker is legitimate. This work provides a machine-learning strategy for predicting user click fraud, which will enable us to distinguish between fraudulent and legitimate clicks and, therefore, identify fraudulent users from legitimate ones. We have used KNN, SVC, and Random Forest models for this purpose.

Association of single-click radiomic classifier with response and prognosis in non-small cell lung cancers (NSCLC) treated with immune checkpoint inhibitors.

8574 Background: Radiomics has shown promise to non-invasively phenotype disease and address the limitations of extant biomarkers (e.g. PD for immune checkpoint inhibitors (ICI) in cancers, such as NSCLC. However, considerable barriers to the clinical adoption of these tools remain, such as their dependence on precise annotation of tumor extent by experienced clinical users. Here, we demonstrate a radiomic solution that requires only a single user mouse click within one or more target lesions on a baseline CT scan, to contour tumors in 3D and generate a patient-level radiomic prediction of response and outcome in ICI treated NSCLC patients. Methods: 1778 CT scans from 1261 patients were used to develop and validate an interactive, semi-automated tool for predicting ICI outcomes in NSCLC patients prior to therapy. A user click based deep learning contouring model was trained and validated on 1146 patients, then used to create annotations for radiomic analysis. A least absolute shrinkage and selection operator (LASSO) Cox proportional hazards model was utilized to select features associated with post-ICI overall survival (OS) and derive a radiomic risk score within a training cohort (n=74) that can separate patients into high and low risk groups. The model was tested on held out pre-treatment CTs of 41 ICI recipients from 2 institutions for association with OS, progression-free survival (PFS), and objective response (OR). Results: A total of 77 lesions were identified and segmented within the testing set. Average volume per lesion was 54.10 mL and per patient was 101.60 mL. OR was observed in 48.70% of patients. A threshold of -0.31 defining high and low radiomic risk groups was chosen based on optimal separation within the training set (HR=2.59 [95% 1.48~4.50], p=0.0009). Radiomic risk groups significantly stratified patients by OS (C-index=0.64, HR=3.03 [95% 1.15~8.02], p=0.03) and PFS (C-index=0.59, HR=3.20 [95% 1.13~9.10], p=0.03). Radiomic IO risk group was independently prognostic of clinical variables (Table 1) and further predicted ICI response with AUC=0.74 [95% 0.71-0.78]. Conclusions: From a single click in target lesions, our model was able to predict response and prognosis of ICI recipients from a baseline radiology scan. Additional multi-site validation and prospective evaluation will assess the value of the radiomics classifier as a decision support tool in the clinic. [Table: see text]