Share This Author
Bid optimizing and inventory scoring in targeted online advertising
- Claudia Perlich, Brian Dalessandro, Rod Hook, Ori Stitelman, Troy Raeder, F. Provost
- 12 August 2012
This paper presents a bid-optimization approach that is implemented in production at Media6Degrees for bidding on these advertising opportunities at an appropriate price and combines several supervised learning algorithms, as well as second price auction theory, to determine the correct price.
Causally motivated attribution for online advertising
- Brian Dalessandro, Claudia Perlich, Ori Stitelman, F. Provost
- Computer Science, BusinessADKDD '12
- 12 August 2012
A causally motivated methodology for conversion attribution in online advertising campaigns is presented and it is argued that in cases where causal assumptions are violated, these approximate methods can be interpreted as variable importance measures.
A Market-Based Framework for Bankruptcy Prediction
We estimate probabilities of bankruptcy for 5,784 industrial firms in the period 1988-2002 in a model where common equity is viewed as a down-and-out barrier option on the firm's assets. Asset values…
Spatial-temporal causal modeling for climate change attribution
This work develops a novel method to infer causality from spatial-temporal data, as well as a procedure to incorporate extreme value modeling into this method in order to address the attribution of extreme climate events, such as heatwaves.
Tree Induction Vs Logistic Regression: A Learning Curve Analysis
A large-scale experimental comparison of logistic regression and tree induction is presented, assessing classification accuracy and the quality of rankings based on class-membership probabilities, and a learning-curve analysis is used to examine the relationship of these measures to the size of the training set.
Aggregation-based feature invention and relational concept classes
It is demonstrated empirically on a noisy business domain that more-complex aggregation methods can increase generalization performance and constructing features using target-dependent aggregations can transform relational prediction tasks so that well-understood feature-vector-based modeling algorithms can be applied successfully.
Leakage in data mining: formulation, detection, and avoidance
It is shown that it is possible to avoid leakage with a simple specific approach to data management followed by what the authors call a learn-predict separation, and several ways of detecting leakage when the modeler has no control over how the data have been collected.
Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining
- Sofus A. Macskassy, Claudia Perlich, J. Leskovec, W. Wang, R. Ghani
- Computer ScienceKDD
- 24 August 2014
It is the authors' great pleasure to welcome you to the 20th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), which this year is partnering with Bloomberg to emphasize the theme of Data Science for Social Good.
Evaluating and Optimizing Online Advertising: Forget the Click, but There Are Good Proxies
A detailed treatment of proxy modeling, which is based on the identification of a suitable alternative (proxy) target variable when data on the true objective is in short supply (or even completely nonexistent), is presented.