• Corpus ID: 13240553

Ensemble of Collaborative Filtering and Feature Engineered Models for Click Through Rate Prediction

@inproceedings{Tscher2012EnsembleOC,
  title={Ensemble of Collaborative Filtering and Feature Engineered Models for Click Through Rate Prediction},
  author={Andreas T{\"o}scher and Michael Jahrer and Jeong-Yoon Lee},
  year={2012}
}
The challenge for Track 2 of the KDD Cup 2012 competition was to predict the click-through rate (CTR) of web advertisements given information about the ad, the query and the user. Our solution comprised an ensemble of models, combined using an artificial neural network. We built collaborative filters, probability models, and feature engineered models to predict CTRs. In addition, we developed a few models which directly optimized AUC, including the collaborative filters and ANN models. These… 
kaggle2.blob.core.windows.net
19 Citations
Highly Influential Citations
1
Background Citations
5
Methods Citations
9
Results Citations
1

19 Citations

Hybrid Models of Factorization Machines with Neural Networks and Their Ensembles for Click-through Rate Prediction
  • Hakan Toğuç, R. Kuzu
  • Computer Science
    2020 5th International Conference on Computer Science and Engineering (UBMK)
  • 2020
TLDR
The best ensemble model is implemented on a real digital campaign to show both prediction performance and the financial outcome of the CTR prediction comparatively, and is based on the serial ensem-bling of hybrid models via Gradient Boosting Algorithm.
A New Approach for Advertising CTR Prediction Based on Deep Neural Network via Attention Mechanism
TLDR
The method exploits dimension reduction based on decomposition, takes advantage of the attention mechanism in neural network modelling, and improves FM to make feature interactions contribute differently to the prediction.
Meta-Wrapper: Differentiable Wrapping Operator for User Interest Selection in CTR Prediction
TLDR
This paper proposes a novel approach under the framework of the wrapper method that efficiencies the wrapper-based feature selection, and achieves better resistance to overfitting, and extensive experiments manifest the superiority of the method in boosting the performance of CTR prediction.
Field-aware Factorization Machines for CTR Prediction
TLDR
This paper establishes FFMs as an effective method for classifying large sparse data including those from CTR prediction, and proposes efficient implementations for training FFMs and comprehensively analyze FFMs.
Research on CTR Prediction Based on Deep Learning
TLDR
The method exploits dimension reduction based on decomposition and combines the power of field-aware factorization machines and deep learning to portray the nonlinear associated relationship of data to solve the sparse feature learning problem.
Improving Ad Click Prediction by Considering Non-displayed Events
TLDR
This paper proposes a novel framework for counterfactual CTR prediction by considering not only displayed events but also non-displayed events and compares this framework against state-of-the-art conventional CTR models and existingcounterfactual learning approaches.
CTR Prediction Models Considering the Dynamics of User Interest
TLDR
A deep-based dynamic interest perception network (DIPN) that can trace both positive interest and negative interest is proposed and the experimental results demonstrate that compared with state-of-the-art models, DIPN achieves the highest prediction performance.
A CTR prediction model based on user interest via attention mechanism
TLDR
A novel attentive deep interest-based network model called ADIN that captures the interest sequence in the interest extractor layer, and the auxiliary losses are employed to produce the interest state with the deep supervision.
DexDeepFM: Ensemble Diversity Enhanced Extreme Deep Factorization Machine Model
TLDR
An ensemble diversity enhanced extreme deep factorization machine model (DexDeepFM) is proposed, which designs the ensemble diversity measure in each hidden layer and considers both ensemble diversity and prediction accuracy in the objective function.
E-Commerce Item Recommendation Based on Field-aware Factorization Machine
TLDR
The team Random Walker's approach to this challenge, which won the 3rd place in the RecSys 2015 contest, consists of casting the top-N recommendation task into a binary classification problem and extracting original features from the raw data using field-aware factorization machines and gradient boosting decision tree.
...
...

References

SHOWING 1-10 OF 17 REFERENCES
Response prediction using collaborative filtering with hierarchies and side-information
TLDR
This paper shows how response prediction can be viewed as a problem of matrix completion, and proposes to solve it using matrix factorization techniques from collaborative filtering (CF), and shows how this factorization can be seamlessly combined with explicit features or side-information for pages and ads, which let us combine the benefits of both approaches.
Collaborative Filtering Ensemble for Ranking
TLDR
The solution of the team "commendo" on the Track2 dataset of the KDD Cup 2011 Dror et al..
Predicting Ads’ Click-Through Rate with Decision Rules
TLDR
This paper builds upon a large data set with real ad clicks and impressions acquired thanks to Microsoft’s Beyond Search program, and presents an algorithm learning an ensemble of decision rules that can be used for predicting the CTR for unseen ads and giving recommendations to improve ads’ quality.
Learning User Behaviors for Advertisements Click Prediction
TLDR
Several machine learning algorithms including conditional random fields (CRF), support vector machines (SVM), decision tree (DT) and backpropagation neural networks (BPN) are developed to learn user's click behaviors from advertisement search and click logs to study the impact of feature selection algorithms on the prediction models.
Identification of factors predicting clickthrough in Web searching using neural network analysis
TLDR
This research uses a neural network to detect the significant influence of searching characteristics on future user clickthrough, and shows that high occurrences of query reformulation, lengthy searching duration, longer query length, and the higher ranking of prior clicked links correlate positively with future clickthrough.
Predicting click through rate for job listings
TLDR
This work learns regression models using features of the job, optional click history of job, features of "related" jobs, and shows that their models predict CTR much better than predicting avg.
Predicting clicks: estimating the click-through rate for new ads
TLDR
This work shows that it can be used to use features of ads, terms, and advertisers to learn a model that accurately predicts the click-though rate for new ads, and shows that using this model improves the convergence and performance of an advertising system.
Improving regularized singular value decomposition for collaborative filtering
TLDR
Different efficient collaborative filtering techniques and a framework for combining them to obtain a good prediction are described, predicting users’ preferences for movies with error rate 7.04% better on the Netflix Prize dataset than the reference algorithm Netflix Cinematch.
Greedy function approximation: A gradient boosting machine.
TLDR
A general gradient descent boosting paradigm is developed for additive expansions based on any fitting criterion, and specific algorithms are presented for least-squares, least absolute deviation, and Huber-M loss functions for regression, and multiclass logistic likelihood for classification.
Optimising area under the ROC curve using gradient descent
This paper introduces RankOpt, a linear binary classifier which optimises the area under the ROC curve (the AUC). Unlike standard binary classifiers, RankOpt adopts the AUC statistic as its objective…
...
...