TDCTFIC: A Novel Recommendation Framework Fusing Temporal Dynamics, CNN-Based Text Features and Item Correlation
Summary :
The traditional recommendation system (RS) can learn the potential personal preferences of users and potential attribute characteristics of items through the rating records between users and items to make recommendations.However, for the new items with no historical rating records,the traditional RS usually suffers from the typical cold start problem. Additional auxiliary information has usually been used in the item cold start recommendation,we further bring temporal dynamics,text and relevance in our models to release item cold start.Two new cold start recommendation models TmTx(Time,Text) and TmTI(Time,Text,Item correlation) proposed to solve the item cold start problem for different cold start scenarios.While well-known methods like TimeSVD++ and CoFactor partially take temporal dynamics,comments,and item correlations into consideration to solve the cold start problem but none of them combines these information together.Two models proposed in this paper fused features such as time,text,and relevance can effectively improve the performance under item cold start.We select the convolutional neural network (CNN) to extract features from item description text which provides the model the ability to deal with cold start items.Both proposed models can effectively improve the performance with item cold start.Experimental results on three real-world data set show that our proposed models lead to significant improvement compared with the baseline methods.
- Publication
- IEICE TRANSACTIONS on Information Vol.E102-D No.8 pp.1517-1525
- Publication Date
- 2019/08/01
- Publicized
- 2019/05/14
- Online ISSN
- 1745-1361
- DOI
- 10.1587/transinf.2019EDP7014
- Type of Manuscript
- PAPER
- Category
- Data Engineering, Web Information Systems
Authors
Meng Ting XIONG
Chongqing University
Yong FENG
Chongqing University
Ting WU
Chongqing University
Jia Xing SHANG
Chongqing University
Bao Hua QIANG
Guilin University of Electronic Technology
Ya Nan WANG
Chongqing University
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Meng Ting XIONG, Yong FENG, Ting WU, Jia Xing SHANG, Bao Hua QIANG, Ya Nan WANG, "TDCTFIC: A Novel Recommendation Framework Fusing Temporal Dynamics, CNN-Based Text Features and Item Correlation" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 8, pp. 1517-1525, August 2019, doi: 10.1587/transinf.2019EDP7014.
Abstract: The traditional recommendation system (RS) can learn the potential personal preferences of users and potential attribute characteristics of items through the rating records between users and items to make recommendations.However, for the new items with no historical rating records,the traditional RS usually suffers from the typical cold start problem. Additional auxiliary information has usually been used in the item cold start recommendation,we further bring temporal dynamics,text and relevance in our models to release item cold start.Two new cold start recommendation models TmTx(Time,Text) and TmTI(Time,Text,Item correlation) proposed to solve the item cold start problem for different cold start scenarios.While well-known methods like TimeSVD++ and CoFactor partially take temporal dynamics,comments,and item correlations into consideration to solve the cold start problem but none of them combines these information together.Two models proposed in this paper fused features such as time,text,and relevance can effectively improve the performance under item cold start.We select the convolutional neural network (CNN) to extract features from item description text which provides the model the ability to deal with cold start items.Both proposed models can effectively improve the performance with item cold start.Experimental results on three real-world data set show that our proposed models lead to significant improvement compared with the baseline methods.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019EDP7014/_p
Copy
@ARTICLE{e102-d_8_1517,
author={Meng Ting XIONG, Yong FENG, Ting WU, Jia Xing SHANG, Bao Hua QIANG, Ya Nan WANG, },
journal={IEICE TRANSACTIONS on Information},
title={TDCTFIC: A Novel Recommendation Framework Fusing Temporal Dynamics, CNN-Based Text Features and Item Correlation},
year={2019},
volume={E102-D},
number={8},
pages={1517-1525},
abstract={The traditional recommendation system (RS) can learn the potential personal preferences of users and potential attribute characteristics of items through the rating records between users and items to make recommendations.However, for the new items with no historical rating records,the traditional RS usually suffers from the typical cold start problem. Additional auxiliary information has usually been used in the item cold start recommendation,we further bring temporal dynamics,text and relevance in our models to release item cold start.Two new cold start recommendation models TmTx(Time,Text) and TmTI(Time,Text,Item correlation) proposed to solve the item cold start problem for different cold start scenarios.While well-known methods like TimeSVD++ and CoFactor partially take temporal dynamics,comments,and item correlations into consideration to solve the cold start problem but none of them combines these information together.Two models proposed in this paper fused features such as time,text,and relevance can effectively improve the performance under item cold start.We select the convolutional neural network (CNN) to extract features from item description text which provides the model the ability to deal with cold start items.Both proposed models can effectively improve the performance with item cold start.Experimental results on three real-world data set show that our proposed models lead to significant improvement compared with the baseline methods.},
keywords={},
doi={10.1587/transinf.2019EDP7014},
ISSN={1745-1361},
month={August},}
Copy
TY - JOUR
TI - TDCTFIC: A Novel Recommendation Framework Fusing Temporal Dynamics, CNN-Based Text Features and Item Correlation
T2 - IEICE TRANSACTIONS on Information
SP - 1517
EP - 1525
AU - Meng Ting XIONG
AU - Yong FENG
AU - Ting WU
AU - Jia Xing SHANG
AU - Bao Hua QIANG
AU - Ya Nan WANG
PY - 2019
DO - 10.1587/transinf.2019EDP7014
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2019
AB - The traditional recommendation system (RS) can learn the potential personal preferences of users and potential attribute characteristics of items through the rating records between users and items to make recommendations.However, for the new items with no historical rating records,the traditional RS usually suffers from the typical cold start problem. Additional auxiliary information has usually been used in the item cold start recommendation,we further bring temporal dynamics,text and relevance in our models to release item cold start.Two new cold start recommendation models TmTx(Time,Text) and TmTI(Time,Text,Item correlation) proposed to solve the item cold start problem for different cold start scenarios.While well-known methods like TimeSVD++ and CoFactor partially take temporal dynamics,comments,and item correlations into consideration to solve the cold start problem but none of them combines these information together.Two models proposed in this paper fused features such as time,text,and relevance can effectively improve the performance under item cold start.We select the convolutional neural network (CNN) to extract features from item description text which provides the model the ability to deal with cold start items.Both proposed models can effectively improve the performance with item cold start.Experimental results on three real-world data set show that our proposed models lead to significant improvement compared with the baseline methods.
ER -
English
