15,831 Hits in 4.4 sec

An analysis framework for search sequences

Qiaozhu Mei, Kristina Klinkner, Ravi Kumar, Andrew Tomkins
2009 Proceeding of the 18th ACM conference on Information and knowledge management - CIKM '09  
Using this framework we study a set of fourteen sequence analysis tasks with a range of features and models.  ...  Our framework provides (i) a vocabulary to discuss types of features, models, and tasks, (ii) straightforward feature re-use across problems, (iii) realistic baselines for many sequence analysis tasks  ...  A non-sequential feature views a sequence task as a stand alone task, and does not incorporate global consistency information.  ... 
doi:10.1145/1645953.1646284 dblp:conf/cikm/MeiKKT09 fatcat:e5dybso7xvgdbpwcu2aoexlr7e

XDM: Improving Sequential Deep Matching with Unclicked User Behaviors for Recommender System [article]

Fuyu Lv, Mengxue Li, Tonglei Guo, Changlong Yu, Fei Sun, Taiwei Jin, Wilfred Ng
2022 arXiv   pre-print
However, most existing sequential recommendation models take as input clicked or purchased behavior sequences from user-item interactions.  ...  In this work, we attempt to incorporate and model those unclicked item sequences using a new learning approach in order to explore better sequential recommendation technique.  ...  In this work, we aim to integrate the valuable unclicked item sequences with clicked ones as complete user behaviors into SR models' input to enhance performances of sequential deep matching.  ... 
arXiv:2010.12837v4 fatcat:jspwpvv7ujhsrbw4hx3mhmlzs4

Multi-Scale User Behavior Network for Entire Space Multi-Task Learning [article]

Jiarui Jin, Xianyu Chen, Weinan Zhang, Yuanbo Chen, Zaifan Jiang, Zekun Zhu, Zhewen Su, Yong Yu
2022 arXiv   pre-print
Modelling the user's multiple behaviors is an essential part of modern e-commerce, whose widely adopted application is to jointly optimize click-through rate (CTR) and conversion rate (CVR) predictions  ...  Concretely, we introduce a hierarchical framework, where the lower layer models the user's engagement behaviors while the upper layer estimates the user's satisfaction behaviors.  ...  [31] further incorporates the sequential behavior graph to encode the dependence among the user's multiple behaviors.  ... 
arXiv:2208.01889v2 fatcat:douo6c3y3jcizotfmmfqwlxzha

Behavior Sequence Transformer for E-commerce Recommendation in Alibaba [article]

Qiwei Chen, Huan Zhao, Wei Li, Pipei Huang, Wenwu Ou
2019 arXiv   pre-print
In this paper, we propose to use the powerful Transformer model to capture the sequential signals underlying users' behavior sequences for recommendation in Alibaba.  ...  However, most of these works just concatenate different features, ignoring the sequential nature of users' behaviors.  ...  Therefore, in this work, to address the aforementioned problems facing WDL and DIN, we try to incorporate sequential signal of users' behavior sequences into RS at Taobao.  ... 
arXiv:1905.06874v1 fatcat:du2cee4pnrhnbecvpszfszkj3a

An F-shape Click Model for Information Retrieval on Multi-block Mobile Pages [article]

Jianghao Lin, Lingyue Fu, Weiwen Liu, Ruiming Tang, Weinan Zhang, Rui Zhang, Yong Yu
2022 arXiv   pre-print
Secondly, we propose DAG-structured GRUs and a comparison module to model users' sequential (sequential browsing, block skip) and non-sequential (comparison) behaviors respectively.  ...  Most click models focus on user behaviors towards a single list.  ...  Comparisonbased Click Model (CBCM) [42] further considers non-sequential browsing behaviors (e.g., revisit, compare).  ... 
arXiv:2206.08604v2 fatcat:auwsjnhkuzen5h4dmzxhsojtwu

Feature-level Attentive Neural Model for Session-based Recommendation

Qing Yang, Peicheng Luo, Xinghe Cheng, Ning Li, Jingwei Zhang
2020 IEEE Access  
However, Markov chain-based models do not capture the interactions between non-adjacent click behaviors; therefore, this approach does not capture or distinguish either long-or short-term user interests  ...  A GRU can learn a user's sequential behavior based on their click sequence; then, the collected data can be used to predict the user's next behavior.  ... 
doi:10.1109/access.2020.3010590 fatcat:lq42n4fqonae5ppgfvhrd2ekqi

Computing the Similarity of Sequential Behavior

Christopher W. Myers
2005 Proceedings of the Human Factors and Ergonomics Society Annual Meeting  
This technique offers the promise of solving what Anderson (2002) regarded as the non-determinism problem of modeling behavior at the 100-ms level of behavior.  ...  In this paper a technique is proposed to objectively compare behavioral routines, whether the data are obtained from a human or embodied computational model.  ...  The tool provides a means to incorporate eye gaze positions and saccades, mouse locations and movements, and mouse-clicks into a unified stream of sequential data derived from log files produced by users  ... 
doi:10.1177/154193120504901209 fatcat:txywjs5lobeb5iuwhky5r4qu3e

Time Series Analysis of Clickstream Logs from Online Courses [article]

Yohan Jo, Keith Maki, Gaurav Tomar
2018 arXiv   pre-print
Further, these sequential approaches to click log analysis can be successfully imported to other courses when used with results obtained from graphical model behavior modeling.  ...  Click logs are an important source of information about students' learning behaviors, however current literature has limited understanding of how these behaviors are represented within click logs.  ...  Comparing both sequence-aware and non sequence-aware approaches, we find that approaches which incorporate sequential information outperform those which do not at classifying student performance, and generalize  ... 
arXiv:1809.04177v1 fatcat:la424vawsjekvjl5zxzqb5vvvy

Sequential Click Prediction for Sponsored Search with Recurrent Neural Networks [article]

Yuyu Zhang, Hanjun Dai, Chang Xu, Jun Feng, Taifeng Wang, Jiang Bian, Bin Wang, Tie-Yan Liu
2014 arXiv   pre-print
Compared to traditional methods, this framework directly models the dependency on user's sequential behaviors into the click prediction process through the recurrent structure in RNN.  ...  , what ads she clicked or ignored, and how long she spent on the landing pages of clicked ads, etc.  ...  . • We use Recurrent Neural Networks to model user's click sequence, and successfully incorporate sequential dependency into enhancing the accuracy of click prediction. • We conduct large scale experiments  ... 
arXiv:1404.5772v3 fatcat:cdpdue2qmfgupga3qbycb5i45e

Online Academic Course Performance Prediction using Relational Graph Convolutional Neural Network

Hamid Karimi, Tyler Derr, Jiangtao Huang, Jiliang Tang
2020 Educational Data Mining  
Furthermore, we perform ablation feature analysis on the student behavioral features to better understand the inner workings of DOPE.  ...  To this end, we introduce Deep Online Performance Evaluation (DOPE), which first models the student course relations in an online system as a knowledge graph, then utilizes an advanced graph neural network  ...  The reason for including this method is to evaluate the effectiveness of the way we model sequential behavioral data.  ... 
dblp:conf/edm/KarimiDHT20 fatcat:ukp4p4xthvb77i6gcsv2ksf4fy

Sequential Click Prediction for Sponsored Search with Recurrent Neural Networks

Yuyu Zhang, Hanjun Dai, Chang Xu, Jun Feng, Taifeng Wang, Jiang Bian, Bin Wang, Tie-Yan Liu
Compared to traditional methods, this framework directly models the dependency on user's sequential behaviors into the click prediction process through the recurrent structure in RNN.  ...  , what ads she clicked or ignored, and how long she spent on the landing pages of clicked ads, etc.  ...  ., click or non-click, in logs as the true labels.  ... 
doi:10.1609/aaai.v28i1.8917 fatcat:zkcgb7nj6bff5gtuorw7gyqwda

Personalizing Search Results Using Hierarchical RNN with Query-aware Attention [article]

Songwei Ge, Zhicheng Dou, Zhengbao Jiang, Jian-Yun Nie, Ji-Rong Wen
2019 arXiv   pre-print
However, few studies have taken into account the sequential information underlying previous queries and sessions.  ...  Previous studies extracted information such as past clicks, user topical interests, query click entropy and so on to tailor the original ranking.  ...  Most of them ignored the sequential information contained in historical user behaviors.  ... 
arXiv:1908.07600v1 fatcat:zwoa6blpb5ea5dyy4uxwx6ohwu

Deep Learning for Sequential Recommendation: Algorithms, Influential Factors, and Evaluations [article]

Hui Fang, Danning Zhang, Yiheng Shu, Guibing Guo
2020 arXiv   pre-print
In this view, this survey focuses on DL-based sequential recommender systems by taking the aforementioned issues into consideration.  ...  Specifically,we illustrate the concept of sequential recommendation, propose a categorization of existing algorithms in terms of three types of behavioral sequence, summarize the key factors affecting  ...  [76] proposed a Bayesian personalized ranking model for heterogeneous behavior types (BPRH) that incorporates the target behavior, auxiliary behavior, and negative behavior into a unified model, and  ... 
arXiv:1905.01997v3 fatcat:i7hvdiqjpnaupcq2osrblttb4u

Gumble Softmax For User Behavior Modeling [article]

Weiqi Shao and Xu Chen and Jiashu Zhao and Long Xia and Dawei Yin
2022 arXiv   pre-print
The DIA module make the historical clicked items into a group of item groups and constructs user's dynamic interest representation.  ...  We propose a sequential model with dynamic number of representations for recommendation systems (RDRSR).  ...  recommendation performance. • STAMP [18] is a neural sequential model by incorporating user short-term memories and preferences.  ... 
arXiv:2112.02787v2 fatcat:kfxm3o5wqfabzfs7cq2krggcde

Dynamic attention-integrated neural network for session-based news recommendation

Lemei Zhang, Peng Liu, Jon Atle Gulla
2019 Machine Learning  
In addition, previous work on session-based algorithms only considers user sequence behaviors within current session without incorporating users' historical interests or pointing out users' main purposes  ...  News article semantic embedding, user interests modelling, session-based public behavior mining and an attention scheme that used to learn the attention score of user and item interaction within sessions  ...  -Neural Attentive Recommendation Machine (NARM) 8 : The model incorporates an itemlevel attention mechanism into RNN for capturing both the user's sequential behavior and main purpose in the current session  ... 
doi:10.1007/s10994-018-05777-9 fatcat:vaumwiyy7bbinj47kyguixtqxe
« Previous Showing results 1 — 15 out of 15,831 results