Hybrid scientific article recommendation system with COOT optimization
Today, recommendation systems are everywhere, making a variety of activities considerably more manageable. These systems help users by personalizing their suggestions to their interests and needs. They can propose various goods, including music, courses, articles, agricultural products, fertilizers,...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co. Ltd.
2024-06-01
|
| Series: | Data Science and Management |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666764923000516 |
| _version_ | 1797261764184768512 |
|---|---|
| author | R. Sivasankari J. Dhilipan |
| author_facet | R. Sivasankari J. Dhilipan |
| author_sort | R. Sivasankari |
| collection | DOAJ |
| description | Today, recommendation systems are everywhere, making a variety of activities considerably more manageable. These systems help users by personalizing their suggestions to their interests and needs. They can propose various goods, including music, courses, articles, agricultural products, fertilizers, books, movies, and foods. In the case of research articles, recommendation algorithms play an essential role in minimizing the time required for researchers to find relevant articles. Despite multiple challenges, these systems must solve serious issues such as the cold start problem, article privacy, and changing user interests. This research addresses these issues through the use of two techniques: hybrid recommendation systems and COOT optimization. To generate article recommendations, a hybrid recommendation system integrates features from content-based and graph-based recommendation systems. COOT optimization is used to optimize the results, inspired by the movement of water birds. The proposed method combines a graph-based recommendation system with COOT optimization to increase accuracy and reduce result inaccuracies. When compared to the baseline approaches described, the model provided in this study improves precision by 2.3%, recall by 1.6%, and Mean Reciprocal Rank (MRR) by 5.7%. |
| first_indexed | 2024-04-24T23:46:24Z |
| format | Article |
| id | doaj.art-3aa7d86cea76478f93e2cb2163389bf6 |
| institution | Directory Open Access Journal |
| issn | 2666-7649 |
| language | English |
| last_indexed | 2024-04-24T23:46:24Z |
| publishDate | 2024-06-01 |
| publisher | KeAi Communications Co. Ltd. |
| record_format | Article |
| series | Data Science and Management |
| spelling | doaj.art-3aa7d86cea76478f93e2cb2163389bf62024-03-15T04:44:43ZengKeAi Communications Co. Ltd.Data Science and Management2666-76492024-06-017299107Hybrid scientific article recommendation system with COOT optimizationR. Sivasankari0J. Dhilipan1Corresponding author.; Department of Computer Applications (M.C.A), SRM Institute of Science and Technology-Ramapuram Campus, Ramapuram, Chennai, IndiaDepartment of Computer Applications (M.C.A), SRM Institute of Science and Technology-Ramapuram Campus, Ramapuram, Chennai, IndiaToday, recommendation systems are everywhere, making a variety of activities considerably more manageable. These systems help users by personalizing their suggestions to their interests and needs. They can propose various goods, including music, courses, articles, agricultural products, fertilizers, books, movies, and foods. In the case of research articles, recommendation algorithms play an essential role in minimizing the time required for researchers to find relevant articles. Despite multiple challenges, these systems must solve serious issues such as the cold start problem, article privacy, and changing user interests. This research addresses these issues through the use of two techniques: hybrid recommendation systems and COOT optimization. To generate article recommendations, a hybrid recommendation system integrates features from content-based and graph-based recommendation systems. COOT optimization is used to optimize the results, inspired by the movement of water birds. The proposed method combines a graph-based recommendation system with COOT optimization to increase accuracy and reduce result inaccuracies. When compared to the baseline approaches described, the model provided in this study improves precision by 2.3%, recall by 1.6%, and Mean Reciprocal Rank (MRR) by 5.7%.http://www.sciencedirect.com/science/article/pii/S2666764923000516Recommendation systemCOOT optimizationCitation networkClusteringLong short-term memory (LSTM) |
| spellingShingle | R. Sivasankari J. Dhilipan Hybrid scientific article recommendation system with COOT optimization Data Science and Management Recommendation system COOT optimization Citation network Clustering Long short-term memory (LSTM) |
| title | Hybrid scientific article recommendation system with COOT optimization |
| title_full | Hybrid scientific article recommendation system with COOT optimization |
| title_fullStr | Hybrid scientific article recommendation system with COOT optimization |
| title_full_unstemmed | Hybrid scientific article recommendation system with COOT optimization |
| title_short | Hybrid scientific article recommendation system with COOT optimization |
| title_sort | hybrid scientific article recommendation system with coot optimization |
| topic | Recommendation system COOT optimization Citation network Clustering Long short-term memory (LSTM) |
| url | http://www.sciencedirect.com/science/article/pii/S2666764923000516 |
| work_keys_str_mv | AT rsivasankari hybridscientificarticlerecommendationsystemwithcootoptimization AT jdhilipan hybridscientificarticlerecommendationsystemwithcootoptimization |