Personalized click prediction in sponsored search

@inproceedings{Cheng2010PersonalizedCP,
  title={Personalized click prediction in sponsored search},
  author={Haibin Cheng and Erick Cant{\'u}-Paz},
  booktitle={WSDM '10},
  year={2010}
}
Sponsored search is a multi-billion dollar business that generates most of the revenue for search engines. Predicting the probability that users click on ads is crucial to sponsored search because the prediction is used to influence ranking, filtering, placement, and pricing of ads. Ad ranking, filtering and placement have a direct impact on the user experience, as users expect the most useful ads to rank high and be placed in a prominent position on the page. Pricing impacts the advertisers… 

Figures and Tables from this paper

Estimating ad group performance in sponsored search
TLDR
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.
Relational click prediction for sponsored search
TLDR
This work designs a continuous conditional random fields based model for click prediction, which considers both the features of an ad and its similarity to the surrounding ads, and shows that the model can be effectively learned using maximum likelihood estimation, and can be efficiently inferred due to its closed form solution.
User-Aware Advertisability
TLDR
The concept of user-aware advertisability, which refers to the probability of ad-click on sponsored ads when a specific user submits a query, is introduced and the experimental results show that user preference is of great value for generating user-specific advertisability.
Psychological advertising: exploring user psychology for click prediction in sponsored search
TLDR
This paper proposes modeling user psychological desire in sponsored search according to Maslow's desire theory, which categorizes psychological desire into five levels and each one is represented by a set of textual patterns automatically mined from ad texts.
Exploiting contextual factors for click modeling in sponsored search
TLDR
This is the first paper to examine and model the effects of the above contextual factors in sponsored search, and propose novel click models which can outperform state-of-the-art methods.
Multimedia features for click prediction of new ads in display advertising
TLDR
It is demonstrated that adding multimedia features can significantly improve the accuracy of click prediction for new ads, compared to a state-of-the-art baseline model.
Optimizing Sponsored Search Ranking Strategy by Deep Reinforcement Learning
TLDR
A generic framework to optimize the ranking functions by deep reinforcement learning methods is proposed, allowing adequate exploration of the ranking function parameter space without hurting the performance of the commercial platform.
Click Through Rate Prediction for Display Advertisement
TLDR
A predictive model that generates the click through rate based on different dimensions of ad placement for display advertisements using statistical machine learning regression techniques such as multivariate linear regression (LR), poisson regression (PR) and support vector regression(SVR).
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 16 REFERENCES
Online learning from click data for sponsored search
TLDR
It is found that online multilayer perceptron learning, based on a small set of features representing content similarity of different kinds, significantly outperforms an information retrieval baseline and other learning models, providing a suitable framework for the sponsored search task.
Modeling and predicting user behavior in sponsored search
TLDR
This paper first study user behavior on sponsored search results (i.e., the advertisements displayed by search engines next to the organic results), and compares this behavior to that of organic results, and develops a generative model to mimic trends in observed user activity using a mixture of pareto distributions.
Contextual advertising by combining relevance with click feedback
TLDR
This paper shows how this match between individual ads and the content of the page where the ads are shown can be improved significantly by augmenting the ad-page scoring function with extra parameters from a logistic regression model on the words in the pages and ads.
An experimental comparison of click position-bias models
TLDR
A cascade model, where users view results from top to bottom and leave as soon as they see a worthwhile document, is the best explanation for position bias in early ranks.
Improving Web Search Ranking by Incorporating User Behavior Information
TLDR
It is shown that incorporating user behavior data can significantly improve ordering of top results in real web search setting, improving the accuracy of a competitive web search ranking algorithms by as much as 31% relative to the original performance.
Sponsored search: an overview of the concept, history, and technology
TLDR
This paper conceptualise the sponsored search process as an aspect of information searching, and provides a brief history of sponsored search and an extensive examination of the technology making sponsored search possible.
Keyword generation for search engine advertising using semantic similarity between terms
TLDR
This paper seeks to establish a mathematical formulation of this problem and suggests a method for generation of several terms from a seed keyword by using a web based kernel function to establish semantic similarity between terms.
Syntactic Information Retrieval
TLDR
Experimental results show that this approach to NLP-based IR has the potential to outperform the standard bag of words approach to information retrieval, especially in response to long queries.
A comparison of numerical optimizers for logistic regression
TLDR
This note compares eight different algorithms for computing the maximum a-posteriori parameter estimate and finds the fastest algorithms turn out to be conjugate gradient ascent and quasi-Newton algorithms, which far outstrip Iterative Scaling and its variants.
Some methods for classification and analysis of multivariate observations
The main purpose of this paper is to describe a process for partitioning an N-dimensional population into k sets on the basis of a sample. The process, which is called 'k-means,' appears to give
...
1
2
...