Best Next Preference Prediction Based on LSTM and Multi-level Interactions

Ivett Fuentes; Gonzalo Nápoles; Leticia Arco; Koen Vanhoof

doi:10.1007/978-3-030-82193-7_46

Best Next Preference Prediction Based on LSTM and Multi-level Interactions

Ivett Fuentes^*, Gonzalo Nápoles, Leticia Arco, Koen Vanhoof

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

Abstract

Predict customer buying behavior is an important task for improving direct marketing campaigns, offering the best possible experiences, and providing personalization in the customer journey trip. Improving how models capture the sequential information from transactional data is essential to learn customer buying order and repetitive buying patterns to generate recommendations over time. In this paper, we propose the deep neural network approach DeepCBPP, which models the sequence prediction problem as a multi-class classification problem and takes the LSTM neural network as the base of the training process.

Original language	English
Title of host publication	Intelligent Systems and Applications
Pages	682-699
Number of pages	18
DOIs	https://doi.org/10.1007/978-3-030-82193-7_46
Publication status	Published - 2022
Event	Intelligent Systems Conference (IntelliSys) 2021 - Duration: 2 Sept 2021 → …

Conference

Conference	Intelligent Systems Conference (IntelliSys) 2021
Period	2/09/21 → …

Keywords

CustomeSequence representation
Sequence prediction models
LSTM
Customer buying behavior
Multi-class classification

Access to Document

10.1007/978-3-030-82193-7_46

Cite this

@inproceedings{6dbda3f4ae60447f824ad63f6d0d8bbb,

title = "Best Next Preference Prediction Based on LSTM and Multi-level Interactions",

abstract = "Predict customer buying behavior is an important task for improving direct marketing campaigns, offering the best possible experiences, and providing personalization in the customer journey trip. Improving how models capture the sequential information from transactional data is essential to learn customer buying order and repetitive buying patterns to generate recommendations over time. In this paper, we propose the deep neural network approach DeepCBPP, which models the sequence prediction problem as a multi-class classification problem and takes the LSTM neural network as the base of the training process.",

keywords = "CustomeSequence representation, Sequence prediction models, LSTM, Customer buying behavior, Multi-class classification",

author = "Ivett Fuentes and Gonzalo N{\'a}poles and Leticia Arco and Koen Vanhoof",

year = "2022",

doi = "10.1007/978-3-030-82193-7_46",

language = "English",

pages = "682--699",

booktitle = "Intelligent Systems and Applications",

note = "Intelligent Systems Conference (IntelliSys) 2021 ; Conference date: 02-09-2021",

}

TY - GEN

T1 - Best Next Preference Prediction Based on LSTM and Multi-level Interactions

AU - Fuentes, Ivett

AU - Nápoles, Gonzalo

AU - Arco, Leticia

AU - Vanhoof, Koen

PY - 2022

Y1 - 2022

N2 - Predict customer buying behavior is an important task for improving direct marketing campaigns, offering the best possible experiences, and providing personalization in the customer journey trip. Improving how models capture the sequential information from transactional data is essential to learn customer buying order and repetitive buying patterns to generate recommendations over time. In this paper, we propose the deep neural network approach DeepCBPP, which models the sequence prediction problem as a multi-class classification problem and takes the LSTM neural network as the base of the training process.

AB - Predict customer buying behavior is an important task for improving direct marketing campaigns, offering the best possible experiences, and providing personalization in the customer journey trip. Improving how models capture the sequential information from transactional data is essential to learn customer buying order and repetitive buying patterns to generate recommendations over time. In this paper, we propose the deep neural network approach DeepCBPP, which models the sequence prediction problem as a multi-class classification problem and takes the LSTM neural network as the base of the training process.

KW - CustomeSequence representation

KW - Sequence prediction models

KW - LSTM

KW - Customer buying behavior

KW - Multi-class classification

U2 - 10.1007/978-3-030-82193-7_46

DO - 10.1007/978-3-030-82193-7_46

M3 - Conference contribution

SP - 682

EP - 699

BT - Intelligent Systems and Applications

T2 - Intelligent Systems Conference (IntelliSys) 2021

Y2 - 2 September 2021

ER -

Best Next Preference Prediction Based on LSTM and Multi-level Interactions

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this