Hybrid Deep Neural Networks for Friend Recommendations in Edge Computing Environment
With the rising popularity of social networks and service recommendations, research on new methods of friend recommendation have become a key topic, especially when based on quality-driven resource processing in an edge computing environment. Traditional methods seldom systematically combine static...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8930557/ |
| _version_ | 1818663822906359808 |
|---|---|
| author | Jibing Gong Yi Zhao Shuai Chen Hongfei Wang Linfeng Du Shuli Wang Md Zakirul Alam Bhuiyan Hao Peng Bowen Du |
| author_facet | Jibing Gong Yi Zhao Shuai Chen Hongfei Wang Linfeng Du Shuli Wang Md Zakirul Alam Bhuiyan Hao Peng Bowen Du |
| author_sort | Jibing Gong |
| collection | DOAJ |
| description | With the rising popularity of social networks and service recommendations, research on new methods of friend recommendation have become a key topic, especially when based on quality-driven resource processing in an edge computing environment. Traditional methods seldom systematically combine static attributes (e.g., interests, geographical locations, and common friends), dynamic behaviors (e.g., liking, making comments, forwarding and @), and network structures (e.g., social ties) to recommend a new friend to a target user. Meanwhile, with the advent of deep learning, it has become more challenging to integrate these features into a deep neural network framework for friend recommendation. For example, how do we optimally make use of these features to form a united framework and what type of deep neural network architecture should be introduced into a novel recommendation method in an edge computing environment? In this paper, we propose DFRec++, a hybrid deep neural network framework combining attribute attention and network embeddings to make social friend recommendations with the help of both interactive semantics and contextual enhancement. More specifically, we first utilize the latent dirichlet allocation (LDA) topic model to generate common interest topics between users and compute the similarity of the explicit static attribute vector representation of topics, locations, and common friends. Then we feed dynamic behavior attributes into a convolutional neural network (CNN) to obtain the implicit vector representation of the interactions and context between two users. Subsequently, a multi-attention mechanism is designed to further improve the deep vector representation of the attribute information. Next, the LINE-based network embeddings algorithm is applied to embed the network structure into a low-dimensional vector. Finally, the attribute attention vector and the network embeddings are concatenated to form a deep feature representation, which is subsequently fed to a fully connected neural network (FCNN) to capture the probability of friendship between two users. The output of FCNN indicates the probability of two users becoming friends. We conducted experiments on a real-world Weibo dataset and the results show that DFRec++ outperforms several existing methods. |
| first_indexed | 2024-12-17T05:22:58Z |
| format | Article |
| id | doaj.art-429b055e65af44f78cccd3eaf4c3bc58 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-17T05:22:58Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-429b055e65af44f78cccd3eaf4c3bc582022-12-21T22:01:57ZengIEEEIEEE Access2169-35362020-01-018106931070610.1109/ACCESS.2019.29585998930557Hybrid Deep Neural Networks for Friend Recommendations in Edge Computing EnvironmentJibing Gong0https://orcid.org/0000-0003-4449-5845Yi Zhao1https://orcid.org/0000-0002-0919-867XShuai Chen2https://orcid.org/0000-0003-1753-0115Hongfei Wang3https://orcid.org/0000-0002-2164-4273Linfeng Du4https://orcid.org/0000-0002-4389-5575Shuli Wang5https://orcid.org/0000-0003-3584-146XMd Zakirul Alam Bhuiyan6https://orcid.org/0000-0002-9513-9990Hao Peng7https://orcid.org/0000-0001-7422-630XBowen Du8School of Information Science and Engineering, Yanshan University, Qinhuangdao, ChinaSchool of Information Science and Engineering, Yanshan University, Qinhuangdao, ChinaSchool of Information Science and Engineering, Yanshan University, Qinhuangdao, ChinaSchool of Computer Science and Technology, Beihang University, Beijing, ChinaSchool of Computer Science and Technology, Beihang University, Beijing, ChinaDepartment of Applied Mathematics, Yanshan University, Qinhuangdao, ChinaDepartment of Computer and Information Sciences, Fordham University, New York City, NY, USASchool of Cyber Science and Technology, Beihang University, Beijing, ChinaSchool of Computer Science and Technology, Beihang University, Beijing, ChinaWith the rising popularity of social networks and service recommendations, research on new methods of friend recommendation have become a key topic, especially when based on quality-driven resource processing in an edge computing environment. Traditional methods seldom systematically combine static attributes (e.g., interests, geographical locations, and common friends), dynamic behaviors (e.g., liking, making comments, forwarding and @), and network structures (e.g., social ties) to recommend a new friend to a target user. Meanwhile, with the advent of deep learning, it has become more challenging to integrate these features into a deep neural network framework for friend recommendation. For example, how do we optimally make use of these features to form a united framework and what type of deep neural network architecture should be introduced into a novel recommendation method in an edge computing environment? In this paper, we propose DFRec++, a hybrid deep neural network framework combining attribute attention and network embeddings to make social friend recommendations with the help of both interactive semantics and contextual enhancement. More specifically, we first utilize the latent dirichlet allocation (LDA) topic model to generate common interest topics between users and compute the similarity of the explicit static attribute vector representation of topics, locations, and common friends. Then we feed dynamic behavior attributes into a convolutional neural network (CNN) to obtain the implicit vector representation of the interactions and context between two users. Subsequently, a multi-attention mechanism is designed to further improve the deep vector representation of the attribute information. Next, the LINE-based network embeddings algorithm is applied to embed the network structure into a low-dimensional vector. Finally, the attribute attention vector and the network embeddings are concatenated to form a deep feature representation, which is subsequently fed to a fully connected neural network (FCNN) to capture the probability of friendship between two users. The output of FCNN indicates the probability of two users becoming friends. We conducted experiments on a real-world Weibo dataset and the results show that DFRec++ outperforms several existing methods.https://ieeexplore.ieee.org/document/8930557/Friend recommendationdeep neural networkattribute-specific multi-attention mechanismnetwork embeddingconvolutional neural network |
| spellingShingle | Jibing Gong Yi Zhao Shuai Chen Hongfei Wang Linfeng Du Shuli Wang Md Zakirul Alam Bhuiyan Hao Peng Bowen Du Hybrid Deep Neural Networks for Friend Recommendations in Edge Computing Environment IEEE Access Friend recommendation deep neural network attribute-specific multi-attention mechanism network embedding convolutional neural network |
| title | Hybrid Deep Neural Networks for Friend Recommendations in Edge Computing Environment |
| title_full | Hybrid Deep Neural Networks for Friend Recommendations in Edge Computing Environment |
| title_fullStr | Hybrid Deep Neural Networks for Friend Recommendations in Edge Computing Environment |
| title_full_unstemmed | Hybrid Deep Neural Networks for Friend Recommendations in Edge Computing Environment |
| title_short | Hybrid Deep Neural Networks for Friend Recommendations in Edge Computing Environment |
| title_sort | hybrid deep neural networks for friend recommendations in edge computing environment |
| topic | Friend recommendation deep neural network attribute-specific multi-attention mechanism network embedding convolutional neural network |
| url | https://ieeexplore.ieee.org/document/8930557/ |
| work_keys_str_mv | AT jibinggong hybriddeepneuralnetworksforfriendrecommendationsinedgecomputingenvironment AT yizhao hybriddeepneuralnetworksforfriendrecommendationsinedgecomputingenvironment AT shuaichen hybriddeepneuralnetworksforfriendrecommendationsinedgecomputingenvironment AT hongfeiwang hybriddeepneuralnetworksforfriendrecommendationsinedgecomputingenvironment AT linfengdu hybriddeepneuralnetworksforfriendrecommendationsinedgecomputingenvironment AT shuliwang hybriddeepneuralnetworksforfriendrecommendationsinedgecomputingenvironment AT mdzakirulalambhuiyan hybriddeepneuralnetworksforfriendrecommendationsinedgecomputingenvironment AT haopeng hybriddeepneuralnetworksforfriendrecommendationsinedgecomputingenvironment AT bowendu hybriddeepneuralnetworksforfriendrecommendationsinedgecomputingenvironment |