Deep Learning over Multi-field Categorical Data: A Case Study on User Response Prediction

11 Jan 2016Weinan ZhangTianming DuJun Wang

Predicting user responses, such as click-through rate and conversion rate, are critical in many web applications including web search, personalised recommendation, and online advertising. Different from continuous raw features that we usually found in the image and audio domains, the input features in web space are always of multi-field and are mostly discrete and categorical while their dependencies are little known... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Click-Through Rate Prediction Company* FNN AUC 0.8683 # 2
Click-Through Rate Prediction Company* FNN Log Loss 0.02629 # 2
Click-Through Rate Prediction Criteo FNN AUC 0.7963 # 7
Click-Through Rate Prediction Criteo FNN Log Loss 0.45738 # 8
Click-Through Rate Prediction iPinYou FNN AUC 0.7619 # 4