scholarly journals Session interest model for CTR prediction based on self-attention mechanism

2022 ◽  
Vol 12 (1) ◽  
Author(s):  
Qianqian Wang ◽  
Fang’ai Liu ◽  
Xiaohui Zhao ◽  
Qiaoqiao Tan
Keyword(s):  
User Behavior ◽  
Online Advertising ◽  
Attention Mechanism ◽  
User Behaviors ◽  
Session Layer ◽  
Interest Model ◽  
Highly Correlated ◽  
Click Through Rate ◽  
Gated Recurrent Unit ◽  
Better Than

AbstractClick-through rate prediction, which aims to predict the probability of the user clicking on an item, is critical to online advertising. How to capture the user evolving interests from the user behavior sequence is an important issue in CTR prediction. However, most existing models ignore the factor that the sequence is composed of sessions, and user behavior can be divided into different sessions according to the occurring time. The user behaviors are highly correlated in each session and are not relevant across sessions. We propose an effective model for CTR prediction, named Session Interest Model via Self-Attention (SISA). First, we divide the user sequential behavior into session layer. A self-attention mechanism with bias coding is used to model each session. Since different session interest may be related to each other or follow a sequential pattern, next, we utilize gated recurrent unit (GRU) to capture the interaction and evolution of user different historical session interests in session interest extractor module. Then, we use the local activation and GRU to aggregate their target ad to form the final representation of the behavior sequence in session interest interacting module. Experimental results show that the SISA model performs better than other models.

scholarly journals Deep Session Interest Network for Click-Through Rate Prediction

2019 ◽  
Author(s):  
Yufei Feng ◽  
Fuyu Lv ◽  
Weichen Shen ◽  
Menghan Wang ◽  
Fei Sun ◽  
...  
Keyword(s):  
Recommender Systems ◽  
Target Item ◽  
State Of The Art ◽  
Online Advertising ◽  
Industrial Applications ◽  
Research Topic ◽  
Intrinsic Structure ◽  
User Behaviors ◽  
Click Through Rate ◽  
Behavior Sequences

Click-Through Rate (CTR) prediction plays an important role in many industrial applications, such as online advertising and recommender systems. How to capture users' dynamic and evolving interests from their behavior sequences remains a continuous research topic in the CTR prediction. However, most existing studies overlook the intrinsic structure of the sequences: the sequences are composed of sessions, where sessions are user behaviors separated by their occurring time. We observe that user behaviors are highly homogeneous in each session, and heterogeneous cross sessions. Based on this observation, we propose a novel CTR model named Deep Session Interest Network (DSIN) that leverages users' multiple historical sessions in their behavior sequences. We first use self-attention mechanism with bias encoding to extract users' interests in each session. Then we apply Bi-LSTM to model how users' interests evolve and interact among sessions. Finally, we employ the local activation unit to adaptively learn the influences of various session interests on the target item. Experiments are conducted on both advertising and production recommender datasets and DSIN outperforms other state-of-the-art models on both datasets.

Deep context interaction network based on attention mechanism for click-through rate prediction

2021 ◽  
pp. 1-16
Author(s):  
Ling Yuan ◽  
Zhuwen Pan ◽  
Ping Sun ◽  
Yinzhen Wei ◽  
Haiping Yu
Keyword(s):  
Online Advertising ◽  
Dimensional Space ◽  
Interaction Network ◽  
High Order ◽  
Attention Mechanism ◽  
Feature Interaction ◽  
Feature Interactions ◽  
Low Dimensional ◽  
Click Through Rate ◽  
The Relationship

Click-through rate (CTR) prediction, which aims to predict the probability of a user clicking on an ad, is a critical task in online advertising systems. The problem is very challenging since(1) an effective prediction relies on high-order combinatorial features, and(2)the relationship to auxiliary ads that may impact the CTR. In this paper, we propose Deep Context Interaction Network on Attention Mechanism(DCIN-Attention) to process feature interaction and context at the same time. The context includes other ads in the current search page, historically clicked and unclicked ads of the user. Specifically, we use the attention mechanism to learn the interactions between the target ad and each type of auxiliary ad. The residual network is used to model the feature interactions in the low-dimensional space, and with the multi-head self-attention neural network, high-order feature interactions can be modeled. Experimental results on Avito dataset show that DCIN outperform several existing methods for CTR prediction.

Identifying enhancer–promoter interactions with neural network based on pre-trained DNA vectors and attention mechanism

2019 ◽  
Author(s):  
Zengyan Hong ◽  
Xiangxiang Zeng ◽  
Leyi Wei ◽  
Xiangrong Liu
Keyword(s):  
Cell Lines ◽  
Computational Methods ◽  
Large Scale ◽  
Attention Mechanism ◽  
Global Features ◽  
Local And Global Features ◽  
Dna Vectors ◽  
Genome Scale ◽  
Gated Recurrent Unit ◽  
Better Than

Abstract Motivation Identification of enhancer–promoter interactions (EPIs) is of great significance to human development. However, experimental methods to identify EPIs cost too much in terms of time, manpower and money. Therefore, more and more research efforts are focused on developing computational methods to solve this problem. Unfortunately, most existing computational methods require a variety of genomic data, which are not always available, especially for a new cell line. Therefore, it limits the large-scale practical application of methods. As an alternative, computational methods using sequences only have great genome-scale application prospects. Results In this article, we propose a new deep learning method, namely EPIVAN, that enables predicting long-range EPIs using only genomic sequences. To explore the key sequential characteristics, we first use pre-trained DNA vectors to encode enhancers and promoters; afterwards, we use one-dimensional convolution and gated recurrent unit to extract local and global features; lastly, attention mechanism is used to boost the contribution of key features, further improving the performance of EPIVAN. Benchmarking comparisons on six cell lines show that EPIVAN performs better than state-of-the-art predictors. Moreover, we build a general model, which has transfer ability and can be used to predict EPIs in various cell lines. Availability and implementation The source code and data are available at: https://github.com/hzy95/EPIVAN.

A Speech Enhancement Method Using Attention Mechanism and Gated Recurrent Unit

2021 ◽  
Author(s):  
Kaibei Peng ◽  
Xiaoming Sun ◽  
Haowei Chen ◽  
Zhen He ◽  
Jianrong Wang
Keyword(s):  
Speech Enhancement ◽  
Attention Mechanism ◽  
Enhancement Method ◽  
Gated Recurrent Unit

Relation of blood-free to blood-inclusive postmortem lung water measurements in sheep

1985 ◽  
Vol 59 (2) ◽  
pp. 592-596 ◽  
Author(s):  
J. C. Collins ◽  
J. H. Newman ◽  
N. E. Wickersham ◽  
W. K. Vaughn ◽  
J. R. Snapper ◽  
...  
Keyword(s):  
Extravascular Lung Water ◽  
Weight Ratio ◽  
Lung Homogenate ◽  
Lung Water ◽  
Before And After ◽  
The Mean ◽  
Space Expansion ◽  
Highly Correlated ◽  
Linear Regressions ◽  
Better Than

Our purpose was to see if the postmortem weight ratio of extravascular lung water to blood-free dry lung (blood-free ratio) was related to similar ratios in blood-inclusive lung and in blood. We developed linear regressions of blood-free ratio on ratios for blood-inclusive lung and blood together and for blood-inclusive lung alone for 73 sheep studied under 11 different protocols and for two subgroups of sheep, one with plasma space expansion and the other without expansion. The relation of ratios of blood-free to blood-inclusive lungs was different between the two subgroups. Although all regressions were highly correlated, the fits of the blood-free ratio on ratios for blood-inclusive lung and blood together were better than for blood-inclusive lung alone. The mean error of prediction of extravascular lung water for all sheep was significantly less for the regression of blood-free ratio on ratios for blood and blood-inclusive lung together (11 g) than for blood-inclusive lung alone (18 g). This study shows that weights of lung homogenate and blood samples before and after simple oven drying can be used to provide accurate inexpensive estimates of postmortem extravascular lung water.

Using 10-K Text to Gauge Financial Constraints

2015 ◽  
Vol 50 (4) ◽  
pp. 623-646 ◽  
Author(s):  
Andriy Bodnaruk ◽  
Tim Loughran ◽  
Bill McDonald
Keyword(s):  
Capital Structure ◽  
Financial Constraints ◽  
Securities And Exchange Commission ◽  
Financial Constraint ◽  
Firm Characteristics ◽  
Low Correlation ◽  
Highly Correlated ◽  
Firm Attributes ◽  
The U.S ◽  
Better Than

AbstractMeasuring the extent to which a firm is financially constrained is critical in assessing capital structure. Extant measures of financial constraints focus on macro firm characteristics such as age and size, variables highly correlated with other firm attributes. We parse 10-K disclosures filed with the U.S. Securities and Exchange Commission (SEC) using a unique lexicon based on constraining words. We find that the frequency of constraining words exhibits very low correlation with traditional measures of financial constraints and predicts subsequent liquidity events, such as dividend omissions or increases, equity recycling, and underfunded pensions, better than widely used financial constraint indexes.

Mobile User Behaviors in China

2015 ◽  
pp. 1110-1128
Author(s):  
Minglong Lei ◽  
Weidong Liu ◽  
Yusong Gao ◽  
Tingshao Zhu
Keyword(s):  
Mobile Phones ◽  
User Behavior ◽  
Mobile User ◽  
User Behavior Analysis ◽  
Related Literature ◽  
User Behaviors ◽  
Analysis Process ◽  
Mobile Industry ◽  
The Mainland Of China ◽  
The Relationship

The development of the mobile industry makes it necessary for scholars to study mobile user behaviors in the mainland of China. This article is divided into three main parts after a brief introduction of the current Chinese mobile phone market. The first part is to demonstrate mobile use and its influencing factors in the mainland of China, and then to determine the mostly studied mobile usages among those articles. The second part pays attention to the effect brought by the use of mobile phones, and then checks the relationship between mobile addiction and other social behaviors. The last part is to illustrate the methods employed in the mobile user behavior analysis. After stating the analysis process of user behaviors, the authors attempted to summarize the main features extracted from data mining technology. Finally, the authors put forward some possible directions under the topic of mobile user behavior after careful review of the related literature.

Knowledge Discovery from Online Communities

Data Mining ◽  
2013 ◽  
pp. 1230-1252
Author(s):  
Luca Cagliero ◽  
Alessandro Fiori
Keyword(s):  
Data Mining ◽  
Social Networks ◽  
Social Network ◽  
User Behavior ◽  
Network Models ◽  
Social Model ◽  
User Generated Content ◽  
Growth Trend ◽  
User Behaviors ◽  
Application Systems

This chapter presents an overview of social network features such as user behavior, social models, and user-generated content to highlight the most notable research trends and application systems built over such appealing models and online media data. It first describes the most popular social networks by analyzing the growth trend, the user behaviors, the evolution of social groups and models, and the most relevant types of data continuously generated and updated by the users. Next, the most recent and valuable applications of data mining techniques to social network models and user-generated content are presented. Discussed works address both social model extractions tailored to semantic knowledge inference and automatic understanding of the user-generated content. Finally, prospects of data mining research on social networks are provided as well.

Sign in / Sign up

Export Citation Format

Share Document