Proceedings of the ADKDD'17 2017
DOI: 10.1145/3124749.3124754
|View full text |Cite
|
Sign up to set email alerts
|

Deep & Cross Network for Ad Click Predictions

Abstract: Feature engineering has been the key to the success of many prediction models. However, the process is nontrivial and o en requires manual feature engineering or exhaustive searching. DNNs are able to automatically learn feature interactions; however, they generate all the interactions implicitly, and are not necessarily e cient in learning all types of cross features. In this paper, we propose the Deep & Cross Network (DCN) which keeps the bene ts of a DNN model, and beyond that, it introduces a novel cross n… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

1
710
0
1

Year Published

2019
2019
2024
2024

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 975 publications
(712 citation statements)
references
References 15 publications
1
710
0
1
Order By: Relevance
“…Relative Improvement (RI). It should be noted that a small improvement with respect to AUC is regarded significant for realworld CTR tasks [2,6,15,30]. In order to estimate the relative improvement of our model achieves over the compared models, we here measure RI-AUC and RI-Logloss, which can be formulated as,…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…Relative Improvement (RI). It should be noted that a small improvement with respect to AUC is regarded significant for realworld CTR tasks [2,6,15,30]. In order to estimate the relative improvement of our model achieves over the compared models, we here measure RI-AUC and RI-Logloss, which can be formulated as,…”
Section: Discussionmentioning
confidence: 99%
“…CrossNet (Deep&Cross) [30] (C) is the core of Deep&Cross model, which tries to model feature interactions explicitly by taking outer product of concatenated feature vector at the bit-wise level.…”
Section: Discussionmentioning
confidence: 99%
“…Research on click prediction focus on developing various models, including Logistic Regression (LR) [8,23,27], Polynomial-2 (Poly2) [6], tree-based models [16], tensor-based models [26], Bayesian models [13], Field-aware Factorization Machines (FFM) [18,19], and Field-weighted Factorization Machines (FwFM) [24]. Recently, deep learning for CTR prediction also attracted a lot of research attention [9,14,15,25,29,30,32].…”
Section: Related Workmentioning
confidence: 99%
“…CTR prediction has attracted lots of attention from both academia and industry [2,8,9,17,20]. For example, Logistic Regression (LR) [15] models linear feature importance.…”
Section: Introductionmentioning
confidence: 99%