Efficient and scalable processing of visual property graph queries
Graph-based data structures are extremely important in the field of representing complex relationships across various domains, such as social networks. Among different types of graphs, property graphs stand out because of their ability to store detailed attributes for nodes and edges. However, th...
Saved in:
主要作者: | |
---|---|
其他作者: | |
格式: | Final Year Project |
語言: | English |
出版: |
Nanyang Technological University
2024
|
主題: | |
在線閱讀: | https://hdl.handle.net/10356/175024 |
標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
總結: | Graph-based data structures are extremely important in the field of representing complex
relationships across various domains, such as social networks. Among different types of
graphs, property graphs stand out because of their ability to store detailed attributes for
nodes and edges. However, the increased data volume presents a significant challenge
on how to efficiently query within large-scale property graphs.
This project aims to address these performance issues by proposing a system called
Facraym, which is optimized for blending visual property graph query and processing.
It can efficiently generate query results within large-scale property graphs by utilizing
the delay between user actions. It introduces an adaptive index schema, the Adaptive
Property Candidates (APC) Index, that stores candidates for each query node and edge,
efficiently pruning unavailable candidates to rapidly generate query results.
Additionally, a large-scale property graph dataset called Dummy Social Network
Dataset is generated, which simulates a huge social network. Experiments were con ducted using Facraym to query within this dataset, and the results were compared with
those from a well-known database, Neo4j. Results demonstrate that Facraym reduces
waiting time compared to Neo4j when generating results. |
---|