• Corpus ID: 18217098

Click-Through Prediction for Sponsored Search Advertising with Hybrid Models

  title={Click-Through Prediction for Sponsored Search Advertising with Hybrid Models},
  author={Xingxing Wang and Shijie Lin and Dongying Kong and Liheng Xu and Qiang Yan and Siwei Lai and Liang Wu and Alvin Chin and Guibo Zhu and Heng Gao and Yang Wu and Danny Bickson and Yuanfeng Du and Neng Gong and Chengchun Shu and Shuang Wang and Kang Liu and Shuren Li and Jun Zhao and Fei Tan and Yuanchun Zhou},
In this paper, we report our approach of KDD Cup 2012 track 2 to predicting the click-through rate (CTR) of advertisements. To accurately predict the CTR of an ad is important for commercial search engine companies for deciding the click prices and the order of impressions. We first implemented three existing methods including Online Bayesian Probit Regression (BPR), Support Vector Machine (SVM) and Latent Factor Model (LFM). In order to fully exploit the training set, several Maximum… 

Figures and Tables from this paper

Estimating ad group performance in sponsored search
This paper studies click yield measurement in sponsored search and proposes a novel framework that could directly predict group performance for lists of ads and noticeably outperform existing state-of-the-art approaches.
CTR prediction for contextual advertising: learning-to-rank approach
This paper introduces a click-through rate prediction algorithm based on the learning-to-rank approach that is observed to achieve better results than other baselines in the authors' three metrics.
Ensemble learning methods for pay-per-click campaign management
It is found that the ensemble learning methods were superior classifiers based on a profit per campaign evaluation criterion, and hence increase the profitability of the overall portfolio of campaigns when compared to standard classifiers.
Predicting Search Engine Switching in WSCD 2013 Challenge
This paper describes the solution of GraphLab team that achieves the 4th place for WSCD 2013 Search Engine Switch Detect contest sponsored by Yandex and proposes a two-step ensemble method to blend the authors' individual models in order to fully exploit the dataset and get more accurate result.
Predicting user click behaviour in search engine advertisements
This article surveys the challenge of selecting relevant advertisements that match user needs and have higher probability of attracting users’ attention based on their previous search history, and finds that users with few queries are similar to new users.
Prediction and recommendation in online media
This dissertation designs specific models for online user behavior prediction in various online media: a probabilistic model for personalized tag prediction, a user-tag-specific temporal interests model for tracking users' interests over time in tagging systems, a personalized structure based link prediction model for micro-blogging systems, and a generalized latent factor model and Bayesian treatment for modeling across multiple contexts in online social media.
Analyse et application de la diffusion d'information dans les microblogs. (The analysis and applications of information diffusion in microblogs)
An efficient and unbiased sampling algorithms, Unbiased Sampling method with Dummy Edges (USDE), which reduces strongly the self-sampling probabilities of MHRW and USDSG and a Galton Watson with Killing model which considers all the three important factors including the ephemerality of information, the topology of network and the features of publisher and retweeters accurately.


Personalized click prediction in sponsored search
This paper develops user-specific and demographic-based features that reflect the click behavior of individuals and groups in sponsored search and demonstrates that the personalized models significantly improve the accuracy of click prediction.
Web-Scale Bayesian Click-Through rate Prediction for Sponsored Search Advertising in Microsoft's Bing Search Engine
A new Bayesian click-through rate (CTR) prediction algorithm used for Sponsored Search in Microsoft's Bing search engine is described, based on a probit regression model that maps discrete or real-valued input features to probabilities.
A collaborative filtering approach to ad recommendation using the query-ad click graph
A new technique to determine the relevance of an ad document for a search query using click-through data is proposed, which builds on a collaborative filtering approach to discover new ads related to a query using a click graph.
Predicting clicks: estimating the click-through rate for new ads
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.
Personalized click model through collaborative filtering
A novel personalized click model is put forward to describe the user-oriented click preferences, which applies and extends matrix / tensor factorization from the view of collaborative filtering to connect users, queries and documents together.
Factorization meets the neighborhood: a multifaceted collaborative filtering model
The factor and neighborhood models can now be smoothly merged, thereby building a more accurate combined model and a new evaluation metric is suggested, which highlights the differences among methods, based on their performance at a top-K recommendation task.
Latent Dirichlet Allocation
We propose a generative model for text and other collections of discrete data that generalizes or improves on several previous models including naive Bayes/unigram, mixture of unigrams [6], and
Reciprocal rank fusion outperforms condorcet and individual rank learning methods
Reciprocal Rank Fusion is demonstrated by using RRF to combine the results of several TREC experiments, and to build a meta-learner that ranks the LETOR 3 dataset better than any previously reported method.
Feature-Based Matrix Factorization
Recommender system has been more and more popular and widely used in many applications recently. The increasing information available, not only in quantities but also in types, leads to a big
A support vector method for multivariate performance measures
  • T. Joachims
  • Mathematics, Computer Science
    Künstliche Intell.
  • 2006
An algorithm with which such multivariate SVMs can be trained in polynomial time for large classes of potentially non-linear performance measures, in particular ROCArea and all measures that can be computed from the contingency table are given.