A social information sensitive model for conversational recommender systems

Conversational recommender systems (CRS) facilitate natural language interactions for more effective item suggestions. While these systems show promise, they face challenges in effectively utilizing and integrating informative data with conversation history through semantic fusion. In this study we...

Full description

Saved in:
Bibliographic Details
Main Authors: Abdulaziz Mohammed, Mingwei Zhang, Gehad Abdullah Amran, Husam M. Alawadh, Ruizhe Wang, Amerah Alabrah, Ali A. Al-Bakhrani
Format: Article
Language:English
Published: PeerJ Inc. 2025-08-01
Series:PeerJ Computer Science
Subjects:
Online Access:https://peerj.com/articles/cs-3067.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849227283108200448
author Abdulaziz Mohammed
Mingwei Zhang
Gehad Abdullah Amran
Husam M. Alawadh
Ruizhe Wang
Amerah Alabrah
Ali A. Al-Bakhrani
author_facet Abdulaziz Mohammed
Mingwei Zhang
Gehad Abdullah Amran
Husam M. Alawadh
Ruizhe Wang
Amerah Alabrah
Ali A. Al-Bakhrani
author_sort Abdulaziz Mohammed
collection DOAJ
description Conversational recommender systems (CRS) facilitate natural language interactions for more effective item suggestions. While these systems show promise, they face challenges in effectively utilizing and integrating informative data with conversation history through semantic fusion. In this study we present an innovative framework for extracting social information from conversational datasets by inferring ratings and constructing user-item interaction and user-user relationship graphs. We introduce a social information sensitive semantic fusion (SISSF) method that employs contrastive learning (CL) to bridge the semantic gap between generated social information and conversation history. We evaluated the framework on two public datasets (ReDial and INSPIRED) using both automatic and human evaluation metrics. Our SISSF framework demonstrated significant improvements over baseline models across all metrics. For the ReDial dataset, SISSF achieved superior performance in recommendation tasks (R@1: 0.062, R@50: 0.437) and conversational quality metrics (Distinct-2: 4.223, Distinct-3: 5.595, Distinct-4: 6.155). Human evaluation showed marked improvement in both fluency (1.81) and informativeness (1.63). We observed similar performance gains on the INSPIRED dataset, with notable improvements in recommendation accuracy (R@1: 0.046, R@10: 0.129, R@50: 0.269) and response diversity (Distinct-2: 2.061, Distinct-3: 4.293, Distinct-4: 6.242). The experimental results consistently validate the effectiveness of our approach in both recommendation and conversational tasks. These findings suggest that incorporating social context through CL can significantly improve the personalization and relevance of recommendations in conversational systems.
format Article
id doaj-art-3e28e091e26a4a59a3d522ecc31bc9b4
institution Kabale University
issn 2376-5992
language English
publishDate 2025-08-01
publisher PeerJ Inc.
record_format Article
series PeerJ Computer Science
spelling doaj-art-3e28e091e26a4a59a3d522ecc31bc9b42025-08-23T15:05:16ZengPeerJ Inc.PeerJ Computer Science2376-59922025-08-0111e306710.7717/peerj-cs.3067A social information sensitive model for conversational recommender systemsAbdulaziz Mohammed0Mingwei Zhang1Gehad Abdullah Amran2Husam M. Alawadh3Ruizhe Wang4Amerah Alabrah5Ali A. Al-Bakhrani6College of Software Engineering, Northeastern University, Shenyang, ChinaCollege of Software Engineering, Northeastern University, Shenyang, ChinaDepartment of Management Science and Engineering, Dalian University of Technology, Dalian, ChinaDepartment of English Language, College of Language Sciences, King Saud University, Riyadh, Saudi ArabiaCollege of Software Engineering, Northeastern University, Shenyang, ChinaDepartment of Information Systems, College of Computer and Information Science, King Saud University, Riyadh, Saudi ArabiaCollege of Software Engineering, Dalian University of Technology, Dalian, ChinaConversational recommender systems (CRS) facilitate natural language interactions for more effective item suggestions. While these systems show promise, they face challenges in effectively utilizing and integrating informative data with conversation history through semantic fusion. In this study we present an innovative framework for extracting social information from conversational datasets by inferring ratings and constructing user-item interaction and user-user relationship graphs. We introduce a social information sensitive semantic fusion (SISSF) method that employs contrastive learning (CL) to bridge the semantic gap between generated social information and conversation history. We evaluated the framework on two public datasets (ReDial and INSPIRED) using both automatic and human evaluation metrics. Our SISSF framework demonstrated significant improvements over baseline models across all metrics. For the ReDial dataset, SISSF achieved superior performance in recommendation tasks (R@1: 0.062, R@50: 0.437) and conversational quality metrics (Distinct-2: 4.223, Distinct-3: 5.595, Distinct-4: 6.155). Human evaluation showed marked improvement in both fluency (1.81) and informativeness (1.63). We observed similar performance gains on the INSPIRED dataset, with notable improvements in recommendation accuracy (R@1: 0.046, R@10: 0.129, R@50: 0.269) and response diversity (Distinct-2: 2.061, Distinct-3: 4.293, Distinct-4: 6.242). The experimental results consistently validate the effectiveness of our approach in both recommendation and conversational tasks. These findings suggest that incorporating social context through CL can significantly improve the personalization and relevance of recommendations in conversational systems.https://peerj.com/articles/cs-3067.pdfConversational recommendation systemContrastive learningSemantic fusionSocial recommendationNLP
spellingShingle Abdulaziz Mohammed
Mingwei Zhang
Gehad Abdullah Amran
Husam M. Alawadh
Ruizhe Wang
Amerah Alabrah
Ali A. Al-Bakhrani
A social information sensitive model for conversational recommender systems
PeerJ Computer Science
Conversational recommendation system
Contrastive learning
Semantic fusion
Social recommendation
NLP
title A social information sensitive model for conversational recommender systems
title_full A social information sensitive model for conversational recommender systems
title_fullStr A social information sensitive model for conversational recommender systems
title_full_unstemmed A social information sensitive model for conversational recommender systems
title_short A social information sensitive model for conversational recommender systems
title_sort social information sensitive model for conversational recommender systems
topic Conversational recommendation system
Contrastive learning
Semantic fusion
Social recommendation
NLP
url https://peerj.com/articles/cs-3067.pdf
work_keys_str_mv AT abdulazizmohammed asocialinformationsensitivemodelforconversationalrecommendersystems
AT mingweizhang asocialinformationsensitivemodelforconversationalrecommendersystems
AT gehadabdullahamran asocialinformationsensitivemodelforconversationalrecommendersystems
AT husammalawadh asocialinformationsensitivemodelforconversationalrecommendersystems
AT ruizhewang asocialinformationsensitivemodelforconversationalrecommendersystems
AT amerahalabrah asocialinformationsensitivemodelforconversationalrecommendersystems
AT aliaalbakhrani asocialinformationsensitivemodelforconversationalrecommendersystems
AT abdulazizmohammed socialinformationsensitivemodelforconversationalrecommendersystems
AT mingweizhang socialinformationsensitivemodelforconversationalrecommendersystems
AT gehadabdullahamran socialinformationsensitivemodelforconversationalrecommendersystems
AT husammalawadh socialinformationsensitivemodelforconversationalrecommendersystems
AT ruizhewang socialinformationsensitivemodelforconversationalrecommendersystems
AT amerahalabrah socialinformationsensitivemodelforconversationalrecommendersystems
AT aliaalbakhrani socialinformationsensitivemodelforconversationalrecommendersystems