CD-LLMCARS: Cross Domain Fine-Tuned Large Language Model for Context-Aware Recommender Systems
Recommender systems are essential for providing personalized content across various platforms. However, traditional systems often struggle with limited information, known as the cold start problem, and with accurately interpreting a user's comprehensive preferences, referred to as context...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Open Journal of the Computer Society |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10771726/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841550746519076864 |
|---|---|
| author | Adeel Ashraf Cheema Muhammad Shahzad Sarfraz Usman Habib Qamar Uz Zaman Ekkarat Boonchieng |
| author_facet | Adeel Ashraf Cheema Muhammad Shahzad Sarfraz Usman Habib Qamar Uz Zaman Ekkarat Boonchieng |
| author_sort | Adeel Ashraf Cheema |
| collection | DOAJ |
| description | Recommender systems are essential for providing personalized content across various platforms. However, traditional systems often struggle with limited information, known as the cold start problem, and with accurately interpreting a user's comprehensive preferences, referred to as context. The proposed study, CD-LLMCARS (Cross-Domain fine-tuned Large Language Model for Context-Aware Recommender Systems), presents a novel approach to addressing these issues. CD-LLMCARS leverages the substantial capabilities of the Large Language Model (LLM) Llama 2. Fine-tuning Llama 2 with information from multiple domains can enhance the generation of contextually relevant recommendations that align with a user's preferences in areas such as movies, music, books, and CDs. Techniques such as Low-Rank Adaptation (LoRA) and Half Precision Training (FP16) are both effective and resource-efficient, allowing CD-LLMCARS to perform optimally in cold start scenarios. Extensive testing of CD-LLMCARS indicates outstanding accuracy, particularly in challenging scenarios characterized by limited user history data relevant to the cold start problem. CD-LLMCARS offers precise and pertinent recommendations to users, effectively mitigating the limitations of traditional recommender systems. |
| format | Article |
| id | doaj-art-210f2427241b493f883fa76f89150ba6 |
| institution | Kabale University |
| issn | 2644-1268 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of the Computer Society |
| spelling | doaj-art-210f2427241b493f883fa76f89150ba62025-01-10T00:03:33ZengIEEEIEEE Open Journal of the Computer Society2644-12682025-01-016495910.1109/OJCS.2024.350922110771726CD-LLMCARS: Cross Domain Fine-Tuned Large Language Model for Context-Aware Recommender SystemsAdeel Ashraf Cheema0https://orcid.org/0000-0002-8310-6251Muhammad Shahzad Sarfraz1https://orcid.org/0000-0003-4703-0285Usman Habib2https://orcid.org/0000-0003-4793-6239Qamar Uz Zaman3https://orcid.org/0000-0002-2785-7448Ekkarat Boonchieng4https://orcid.org/0000-0002-7584-1627Department of Computer Science, National University of Computer and Emerging Sciences, Chiniot-Faisalabad Campus, Chiniot, PakistanDepartment of Computer Science, National University of Computer and Emerging Sciences, Chiniot-Faisalabad Campus, Chiniot, PakistanFAST School of Computing, National University of Computer and Emerging Sciences, Islamabad Campus, Islamabad, PakistanDepartment of Software Engineering, National University of Computer and Emerging Sciences, Chiniot-Faisalabad Campus, Chiniot, PakistanDepartment of Computer Science, Faculty of Science, Chiang Mai University, Chiang Mai, ThailandRecommender systems are essential for providing personalized content across various platforms. However, traditional systems often struggle with limited information, known as the cold start problem, and with accurately interpreting a user's comprehensive preferences, referred to as context. The proposed study, CD-LLMCARS (Cross-Domain fine-tuned Large Language Model for Context-Aware Recommender Systems), presents a novel approach to addressing these issues. CD-LLMCARS leverages the substantial capabilities of the Large Language Model (LLM) Llama 2. Fine-tuning Llama 2 with information from multiple domains can enhance the generation of contextually relevant recommendations that align with a user's preferences in areas such as movies, music, books, and CDs. Techniques such as Low-Rank Adaptation (LoRA) and Half Precision Training (FP16) are both effective and resource-efficient, allowing CD-LLMCARS to perform optimally in cold start scenarios. Extensive testing of CD-LLMCARS indicates outstanding accuracy, particularly in challenging scenarios characterized by limited user history data relevant to the cold start problem. CD-LLMCARS offers precise and pertinent recommendations to users, effectively mitigating the limitations of traditional recommender systems.https://ieeexplore.ieee.org/document/10771726/Collaborative filteringcontext-aware recommeder systems (CARS)cross-domain recommender systems (CDRS)large language models (LLMs)prompt engineering |
| spellingShingle | Adeel Ashraf Cheema Muhammad Shahzad Sarfraz Usman Habib Qamar Uz Zaman Ekkarat Boonchieng CD-LLMCARS: Cross Domain Fine-Tuned Large Language Model for Context-Aware Recommender Systems IEEE Open Journal of the Computer Society Collaborative filtering context-aware recommeder systems (CARS) cross-domain recommender systems (CDRS) large language models (LLMs) prompt engineering |
| title | CD-LLMCARS: Cross Domain Fine-Tuned Large Language Model for Context-Aware Recommender Systems |
| title_full | CD-LLMCARS: Cross Domain Fine-Tuned Large Language Model for Context-Aware Recommender Systems |
| title_fullStr | CD-LLMCARS: Cross Domain Fine-Tuned Large Language Model for Context-Aware Recommender Systems |
| title_full_unstemmed | CD-LLMCARS: Cross Domain Fine-Tuned Large Language Model for Context-Aware Recommender Systems |
| title_short | CD-LLMCARS: Cross Domain Fine-Tuned Large Language Model for Context-Aware Recommender Systems |
| title_sort | cd llmcars cross domain fine tuned large language model for context aware recommender systems |
| topic | Collaborative filtering context-aware recommeder systems (CARS) cross-domain recommender systems (CDRS) large language models (LLMs) prompt engineering |
| url | https://ieeexplore.ieee.org/document/10771726/ |
| work_keys_str_mv | AT adeelashrafcheema cdllmcarscrossdomainfinetunedlargelanguagemodelforcontextawarerecommendersystems AT muhammadshahzadsarfraz cdllmcarscrossdomainfinetunedlargelanguagemodelforcontextawarerecommendersystems AT usmanhabib cdllmcarscrossdomainfinetunedlargelanguagemodelforcontextawarerecommendersystems AT qamaruzzaman cdllmcarscrossdomainfinetunedlargelanguagemodelforcontextawarerecommendersystems AT ekkaratboonchieng cdllmcarscrossdomainfinetunedlargelanguagemodelforcontextawarerecommendersystems |