Unsupervised Embedding Learning for Large-Scale Heterogeneous Networks Based on Metapath Graph Sampling

Zhong, Hongwei; Wang, Mingyang; Zhang, Xinyue

doi:10.3390/e25020297

Cited by 7 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The structural feature encoding process aims to learn the scholar's interest vector of structural dimensions derived from the association relationships between academic entities. In our previous work [19], the authors proposed a heterogeneous network representation learning process based on meta-path subgraph sampling. We introduce the process to encode the structural features of scholars' research interests.…”

Section: A Initial Encoding Modulementioning

confidence: 99%

See 1 more Smart Citation

NovSRC: A Novelty-Oriented Scientific Collaborators Recommendation Model

Li,

Wang,

Wang

et al. 2024

IJACSA

View full text Add to dashboard Cite

Collaborator recommendation is a crucial topic in research management. This paper proposes a Novelty-Oriented Scientific Research Collaborator recommendation model (NovSRC). By recommending collaborators under the guidance of novel indicators, NovSRC aims to broaden scholars' research perspectives and facilitate the progress of research innovation. NovSRC utilizes heterogeneous academic networks composed of different academic entities and their relationships to learn vector representations of scholars and quantify their novelty metrics. A weighted academic collaboration network was constructed by measuring the novelty collaboration strength (NCS) among scholars under the novelty index, and based on this network, the final vector representation of scholars under the guidance of novelty characteristics was learned. By calculating the similarity between scholar vectors, NovSRC generates a Top-N recommendation list with a focus on novelty. The experimental results indicate that NovSRC achieved the best recommendation performance. Compared with the baseline models, the recommendation precision of NovSRC has improved by 6.9%, the F1 value has increased by 17.3%, and the novelty collaboration strength among scholars has increased by 3.3%. The analysis of the recommended list shows that compared to the target scholars, scholars recommended by the NovSRC model exhibit a wider distribution of research interests, which confirms that novelty has become a key benchmark factor for scholars seeking collaborators.

show abstract

Section: A Initial Encoding Modulementioning

confidence: 99%

“…Step 3: Based on the influence of heterogeneous neighbors, the fine-tune of the scholar's influence is calculated using formula (19). Specifically, the influence of the paper nodes and journal nodes obtained in step 2 is used to adjust the transition matrix between the scholar nodes in the collaboration subnetwork.…”

Section: 𝐴𝐼𝑆(𝐴mentioning

confidence: 99%

NovSRC: A Novelty-Oriented Scientific Collaborators Recommendation Model

Li,

Wang,

Wang

et al. 2024

IJACSA

View full text Add to dashboard Cite

show abstract

“…One of the main challenges in a dynamic environment is that the size of the graph continuously increases over time. To efficiently manage infinitely increasing data within limited storage space, graph compression is essential, allowing for the efficient use of storage space to accommodate increasing amounts of data [17][18][19][20][21][22]. Techniques that incorporate graph pattern mining methods also exist [23][24][25][26][27][28].…”

Section: Introductionmentioning

confidence: 99%

Graph Stream Compression Scheme Based on Pattern Dictionary Using Provenance

Lee,

Shin,

Choi

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

With recent advancements in network technology and the increasing popularity of the internet, the use of social network services and Internet of Things devices has flourished, leading to a continuous generation of large volumes of graph stream data, where changes, such as additions or deletions of vertices and edges, occur over time. Additionally, owing to the need for the efficient use of storage space and security requirements, graph stream data compression has become essential in various applications. Even though various studies on graph compression methods have been conducted, most of them do not fully reflect the dynamic characteristics of graph streams and the complexity of large graphs. In this paper, we propose a compression scheme using provenance data to efficiently process and analyze large graph stream data. It obtains provenance data by analyzing graph stream data and builds a pattern dictionary based on this to perform dictionary-based compression. By improving the existing dictionary-based graph compression methods, it enables more efficient dictionary management through tracking pattern changes and evaluating their importance using provenance. Furthermore, it considers the relationships among sub-patterns using an FP-tree and performs pattern dictionary management that updates pattern scores based on time. Our experiments show that the proposed scheme outperforms existing graph compression methods in key performance metrics, such as compression rate and processing time.

show abstract

“…One of the main challenges in a dynamic environment is that the size of the graph continuously increases over time. To efficiently manage infinitely increasing data within a limited storage space, graph compression is essential, allowing for the efficient use of storage space to accommodate increasing amounts of data [17][18][19][20][21][22]. Techniques that incorporate graph pattern mining methods also exist [23][24][25][26][27][28].…”

Section: Introductionmentioning

confidence: 99%

Graph Stream Compression Scheme Based on Pattern Dictionary Using Provenance

Lee,

Shin,

Choi

et al. 2024

Preprint

View full text Add to dashboard Cite

With recent advancements in network technology and spread of Internet, use of social network services and Internet of Things devices has flourished, leading to a continuous generation of large volumes of graph stream data, where changes, such as additions or deletions of vertices and edges occur over time. Additionally, owing to the need for efficient use of storage space and security requirements, graph stream data compression has become essential in various applications. Even though various studies on graph compression methods have been conducted, most of them do not fully reflect the dynamic characteristics of graph streams and complexity of large graphs. In this paper, we propose a compression scheme using provenance data to efficiently process and analyze large graph stream data. It obtains provenance data by analyzing graph stream data and builds a pattern dictionary based on this to perform dictionary-based compression. By improving the existing dictionary-based graph compression methods, it enables more efficient dictionary management through tracking pattern changes and evaluating their importance using provenance. Furthermore, it considers the relationships among sub-patterns using an FP-tree and performs pattern dictionary management that updates pattern scores based on time. Our experiments show that the proposed scheme outperforms existing graph compression methods in key performance metrics, such as compression rate and processing time.

show abstract

Unsupervised Embedding Learning for Large-Scale Heterogeneous Networks Based on Metapath Graph Sampling

Cited by 7 publications

References 20 publications

NovSRC: A Novelty-Oriented Scientific Collaborators Recommendation Model

NovSRC: A Novelty-Oriented Scientific Collaborators Recommendation Model

Graph Stream Compression Scheme Based on Pattern Dictionary Using Provenance

Graph Stream Compression Scheme Based on Pattern Dictionary Using Provenance

Contact Info

Product

Resources

About