Conversion Rate Prediction Research Articles

Debiasing the Conversion Rate Prediction Model in the Presence of Delayed Implicit Feedback.

The recommender system (RS) has been widely adopted in many applications, including online advertisements. Predicting the conversion rate (CVR) can help in evaluating the effects of advertisements on users and capturing users' features, playing an important role in RS. In real-world scenarios, implicit rather than explicit feedback data are more abundant. Thus, directly training the RS with collected data may lead to suboptimal performance due to selection bias inherited from the nature of implicit feedback. Methods such as reweighting have been proposed to tackle selection bias; however, these methods omit delayed feedback, which often occurs due to limited observation times. We propose a novel likelihood approach combining the assumed parametric model for delayed feedback and the reweighting method to address selection bias. Specifically, the proposed methods minimize the likelihood-based loss using the multi-task learning method. The proposed methods are evaluated on the real-world Coat and Yahoo datasets. The proposed methods improve the AUC by 5.7% on Coat and 3.7% on Yahoo compared with the best baseline models. The proposed methods successfully debias the CVR prediction model in the presence of delayed implicit feedback.

Online Conversion Rate Prediction via Multi-Interval Screening and Synthesizing under Delayed Feedback

Due to the widespread adoption of the cost-per-action(CPA) display strategy that demands a real-time conversion rate prediction(CVR), delayed feedback is becoming one of the major challenges in online advertising. As the true labels of a significant quantity of samples are only available after long delays, the observed training data are usually biased, harming the performance of models. Recent studies show integrating models with varying waiting windows to observe true labels is beneficial, but the aggregation framework remains far from reaching a consensus. In this work, we propose the Multi-Interval Screening and Synthesizing model (MISS for short) for online CVR prediction. We first design a multi-interval screening model with various output heads to produce accurate and distinctive estimates. Then a light-weight synthesizing model with an assembled training pipeline is applied to thoroughly exploit the knowledge and relationship among heads, obtaining reliable predictions. Extensive experiments on two real-world advertising datasets validate the effectiveness of our model.

Click-through conversion rate prediction model of book e-commerce platform based on feature combination and representation

The prediction of click-through conversion rates has always been an integral focus within e-commerce platforms. This paper proposes a novel prediction model that leverages feature combination and representation learning techniques, centering around three crucial elements of the Buyer-Book-Activity (BBA) e-commerce platform. Specifically, our approach considers the object characteristics of book e-commerce platforms. Firstly, we create a comprehensive three-part BBA diagram encompassing user-book-marketing activities and thoroughly describe the book e-commerce platform’s information space. To construct a comprehensive feature space, we employ a factor decomposition method, enabling us to combine and mine large-scale information from the BBA. In particular, we introduce book2tag text features to capture textual information’s hidden richness and decompose it. Secondly, to represent the overall information space of the book e-commerce platform, we propose the BBA2vec representation learning method. This method captures the complexity and diversity of the BBA’s effective data feature space. This is crucial as feature combinations alone may not fully capture the diversity of e-commerce features. Finally, a deep neural network is used to predict click-through conversion rates by combining feature combination and feature representation. The experimental results show that the model accurately predicts the bookstore e-commerce platform’s click-purchase conversion rate.

Multi-Task Deep Learning with Task Attention for Post-Click Conversion Rate Prediction

Online advertising has gained much attention on various platforms as a hugely lucrative market. In promoting content and advertisements in real life, the acquisition of user target actions is usually a multi-step process, such as impression→click→conversion, which means the process from the delivery of the recommended item to the user’s click to the final conversion. Due to data sparsity or sample selection bias, it is difficult for the trained model to achieve the business goal of the target campaign. Multi-task learning, a classical solution to this problem, aims to generalize better on the original task given several related tasks by exploiting the knowledge between tasks to share the same feature and label space. Adaptively learned task relations bring better performance to make full use of the correlation between tasks. We train a general model capable of capturing the relationships between various tasks on all existing active tasks from a meta-learning perspective. In addition, this paper proposes a Multi-task Attention Network (MAN) to identify commonalities and differences between tasks in the feature space. The model performance is improved by explicitly learning the stacking of task relationships in the label space. To illustrate the effectiveness of our method, experiments are conducted on Alibaba Click and Conversion Prediction (Ali-CCP) dataset. Experimental results show that the method outperforms the state-of-the-art multi-task learning methods.

Research on the Effect of Online Marketing Based on Multimodel Fusion and Artificial Intelligence in the Context of Big Data

Traditional online marketing methods use a single model to predict the advertising conversion rate, but the prediction results are not accurate, and users are not satisfied with the recommendation results. Therefore, this paper proposes an online marketing method based on multimodel fusion and artificial intelligence algorithms under the background of big data. First, it introduces big data technology and analyzes the characteristics of network advertising marketing model (RTB). Second, combined with multitask learning and fusion technology to improve the single model in advertising conversion rate prediction effect, prediction results to further improve the accuracy of results. Then, tF-IDF technology in artificial intelligence algorithm is used to measure the importance of advertising words in online marketing and calculate the contribution degree. Finally, according to XGBoost technology, the multitask fusion model of online marketing effect is classified. Experiments are used to analyze the effect of online marketing. Experimental results show that the proposed method can improve the accuracy of advertising conversion rate prediction and online sales of goods.

Conversion Rate Prediction Based on Text Readability Analysis of Landing Pages.

Digital marketing has been extensively researched and developed remarkably rapidly over the last decade. Within this field, hundreds of scientific publications and patents have been produced, but the accuracy of prediction technologies leaves much to be desired. Conversion prediction remains a problem for most marketing professionals. In this article, the authors, using a dataset containing landing pages content and their conversions, show that a detailed analysis of text readability is capable of predicting conversion rates. They identify specific features that directly affect conversion and show how marketing professionals can use the results of this work. In their experiments, the authors show that the applied machine learning approach can predict landing page conversion. They built five machine learning models. The accuracy of the built machine learning model using the SVM algorithm is promising for its implementation. Additionally, the interpretation of the results of this model was conducted using the SHAP package. Approximately 60% of purchases are made by nonmembers, and this paper may be suitable for the cold-start problem.

Capturing Delayed Feedback in Conversion Rate Prediction via Elapsed-Time Sampling

Conversion rate (CVR) prediction is one of the most critical tasks for digital display advertising. Commercial systems often require to update models in an online learning manner to catch up with the evolving data distribution. However, conversions usually do not happen immediately after user clicks. This may result in inaccurate labeling, which is called delayed feedback problem. In previous studies, delayed feedback problem is handled either by waiting positive label for a long period of time, or by consuming the negative sample on its arrival and then insert a positive duplicate when conversion happens later. Indeed, there is a trade-off between waiting for more accurate labels and utilizing fresh data, which is not considered in existing works. To strike a balance in this trade-off, we propose Elapsed-Time Sampling Delayed Feedback Model (ES-DFM), which models the relationship between the observed conversion distribution and the true conversion distribution. Then we optimize the expectation of true conversion distribution via importance sampling under the elapsed-time sampling distribution. We further estimate the importance weight for each instance, which is used as the weight of loss function in CVR prediction. To demonstrate the effectiveness of ES-DFM, we conduct extensive experiments on a public data and a private industrial dataset. Experimental results confirm that our method consistently outperforms the previous state-of-the-art results.

Attentive capsule network for click-through rate and conversion rate prediction in online advertising

Estimating Click-through Rate (CTR) and Conversion Rate (CVR) are two essential user response prediction tasks in computing advertising and recommendation systems. The mainstream methods map sparse, high-dimensional categorical features (e.g., user id, item id) into low-dimensional representations with neural networks. Although they have achieved significant advancement in recent years, how to capture user’s diverse interests effectively from past behaviors is still challenging. Recently some works try using attention-based methods to learn the representation from user behavior history adaptively. However, it is insufficient to capture the diversity of user’s interests. As a step forward to improve this goal, we propose a method named Attentive Capsule Network (ACN). It uses Transformers for feature interaction and leverages capsule networks to capture multiple interests from user behavior history. To precisely obtain sequence representation related to the current advertisement, we further design a modified dynamic routing algorithm integrating with an attention mechanism. Experimental results on real-world datasets demonstrate the effectiveness of our proposed ACN with significant improvement over state-of-the-art approaches. Moreover, it also offers good explainability when extracting diverse interest points of users from behavior history.

Thermoelectrical-based fuel adaptability analysis of solid oxide fuel cell system and fuel conversion rate prediction

Solid oxide fuel cell (SOFC) has attracted more and more attention for its advantages of high efficiency, low emission, and strong fuel adaptability. However, the fuel adaptability also brings issues for thermoelectric cooperative optimization of SOFC system because different fuel compositions can result in different thermal stresses and system performances. In this paper, the temperature distribution and electrical characteristics of a 1 kW external steam-reforming SOFC system at different power stages are studied based on the 200-hour experimental data obtained from the system fed by hydrogen and methane respectively. The thermoelectric characteristics of SOFC system using hydrogen and methane as fuels with the same output power are further compared. Furthermore, the influences of methane conversion rate (MCR) on the thermoelectric characteristics in SOFC system are analyzed through a mathematical SOFC system model and a Gaussian process regression (GPR) model is constructed to predict the MCR. The results show that, the stack voltage of the SOFC system fed by methane (CH4-SOFC) is lower than that of the SOFC system fed by hydrogen (H2-SOFC) by 0.59–4.35 V for the hydrogen concentration at the anode is relatively low in the CH4-SOFC system. Meanwhile, the electrical efficiency of the CH4-SOFC system is lower than that of the H2-SOFC system by 0.85–6.08% in the power range of 300–600 W. In terms of thermal characteristics, the stack temperature gradient on the fuel direction of the CH4-SOFC system is much larger than that of the H2-SOFC system. Different methane conversion rates (MCRs) in the CH4-SOFC system can shift the stack temperature distribution significantly, and properly increasing the MCR can effectively increase the stack voltage and electrical efficiency. Finally, the GPR model can accurately predict the MCR and the MAPE is 0.030%, which can prevent the deformation and fracture of cells caused by the excessive temperature gradient in the stack when the MCR is too low. This work lays a foundation for system control strategy design in the future, especially SOFC stack thermal management, improving the fuel adaptability of the system and ensuring the safety of the stack temperature.

A New Information-Theoretic Method for Advertisement Conversion Rate Prediction for Large-Scale Sparse Data Based on Deep Learning.

With the development of online advertising technology, the accurate targeted advertising based on user preferences is obviously more suitable both for the market and users. The amount of conversion can be properly increased by predicting the user’s purchasing intention based on the advertising Conversion Rate (CVR). According to the high-dimensional and sparse characteristics of the historical behavior sequences, this paper proposes a LSLM_LSTM model, which is for the advertising CVR prediction based on large-scale sparse data. This model aims at minimizing the loss, utilizing the Adaptive Moment Estimation (Adam) optimization algorithm to mine the nonlinear patterns hidden in the data automatically. Through the experimental comparison with a variety of typical CVR prediction models, it is found that the proposed LSLM_LSTM model can utilize the time series characteristics of user behavior sequences more effectively, as well as mine the potential relationship hidden in the features, which brings higher accuracy and trains faster compared to those with consideration of only low or high order features.

基于组合响应预测模型的转换率预测

基于组合响应预测模型的转换率预测

Multi-Level Deep Cascade Trees for Conversion Rate Prediction in Recommendation System

Developing effective and efficient recommendation methods is very challenging for modern e-commerce platforms. Generally speaking, two essential modules named “ClickThrough Rate Prediction” (CTR) and “Conversion Rate Prediction” (CVR) are included, where CVR module is a crucial factor that affects the final purchasing volume directly. However, it is indeed very challenging due to its sparseness nature. In this paper, we tackle this problem by proposing multiLevel Deep Cascade Trees (ldcTree), which is a novel decision tree ensemble approach. It leverages deep cascade structures by stacking Gradient Boosting Decision Trees (GBDT) to effectively learn feature representation. In addition, we propose to utilize the cross-entropy in each tree of the preceding GBDT as the input feature representation for next level GBDT, which has a clear explanation, i.e., a traversal from root to leaf nodes in the next level GBDT corresponds to the combination of certain traversals in the preceding GBDT. The deep cascade structure and the combination rule enable the proposed ldcTree to have a stronger distributed feature representation ability. Moreover, inspired by ensemble learning, we propose an Ensemble ldcTree (E-ldcTree) to encourage the model’s diversity and enhance the representation ability further. Finally, we propose an improved Feature learning method based on EldcTree (F-EldcTree) for taking adequate use of weak and strong correlation features identified by pretrained GBDT models. Experimental results on off-line data set and online deployment demonstrate the effectiveness of the proposed methods.

Response prediction via integration of heterogeneous information

In recent years, the tensor factorization model has been used to model complicated feature interactions involving multiple aspects, such as the user, publisher, and advertiser, for response prediction in real-time bidding. Among numerous challenges, data sparsity and cold start problems always bother researchers, particularly for ad conversion rate prediction. Such problems in prediction become difficult if only one or several types of information are considered. All types of heterogeneous information must be simultaneously integrated to address these problems. This paper proposes an availability solution for integrating heterogeneous information in the tensor factorization model to efficiently alleviate data sparsity and cold start problems. It proposes different integration strategies and implementation methods for various types of information depending on their property, category, structure, form, and function. This solution efficiently alleviates data sparsity and cold start problems, and enhances the prediction reliability and precision for the tensor factorization model in real-time bidding systems. Finally, this solution achieves a significant improvement in response prediction compared to baselines methods on the selection datasets.

Additional Multi-Touch Attribution for Online Advertising

Multi-Touch Attribution studies the effects of various types of online advertisements on purchase conversions. It is a very important problem in computational advertising, as it allows marketers to assign credits for conversions to different advertising channels and optimize advertising campaigns. In this paper, we propose an additional multi-touch attribution model (AMTA) based on two obvious assumptions: (1) the effect of an ad exposure is fading with time and (2) the effects of ad exposures on the browsing path of a user are additive.AMTA borrows the techniques from survival analysis and uses the hazard rate to measure the influence of an ad exposure. In addition, we both take the conversion time and the intrinsic conversion rate of users into consideration.Experimental results on a large real-world advertising dataset illustrate that the our proposed method is superior to state-of-the-art techniques in conversion rate prediction and the credit allocation based on AMTA is reasonable.

Time-aware conversion prediction

The importance of product recommendation has been well recognized as a central task in business intelligence for e-commerce websites. Interestingly, what has been less aware of is the fact that different products take different time periods for conversion. The “conversion” here refers to actually a more general set of pre-defined actions, including for example purchases or registrations in recommendation and advertising systems. The mismatch between the product’s actual conversion period and the application’s target conversion period has been the subtle culprit compromising many existing recommendation algorithms. The challenging question: what products should be recommended for a given time period to maximize conversion—is what has motivated us in this paper to propose a rank-based time-aware conversion prediction model (rTCP), which considers both recommendation relevance and conversion time. We adopt lifetime models in survival analysis to model the conversion time and personalize the temporal prediction by incorporating context information such as user preference. A novel mixture lifetime model is proposed to further accommodate the complexity of conversion intervals. Experimental results on two real-world data sets illustrate the high goodness of fit of our proposed model rTCP and demonstrate its effectiveness in time-aware conversion rate prediction for advertising and product recommendation.