[WIP] RelationService: node ranking for heterogeneous social graph

[BinaryHB0916]

Description

RelationService aggregates all identities by recording all Web2 and Web3 platform connections.

Based on GraphDB, we provide graph solutions for on-chain and off-chain identity relational aggregation.

RelationService collects other third-party "identity <-> identity" relationships, such as "I" holding my own Avatar, while also holding some asymmetric cryptography ID or Web2 identity (account). "I" have a public verifiable two-way binding between Avatar and identities. "I" own a decentralized digital identity, the structure in RelationService is called "Identity Graph". Identity of users may interact (transfer/follow/message). These connections generate a powerful "Open Social Graph".

The Social Graph is a large and heterogeneous graph, and its size makes it difficult to understand the essential information it contains.

As a result, "Identity Graph" is being compressed. It can be used to improve visualization, understand the graph's high-level structure, or as a pre-processing step for other data mining algorithms.

PageRank, one of the most widely used ranking algorithms, was created to rank websites in search engine results. The algorithm acts on unipartite directed networks and builds on the circular idea "A node is important if it is pointed by other important nodes".

The importance of a node in relation to other network elements is determined by node ranking, which is a critical problem in networks.

Node ranking is also important for enabling Social d/App and AI analytics in our product.

What we‘re doing:

• provide an analytic tool which developers can handle rich graph data conveniently
• provide a set of data analysis algorithms to empower social dApp, market analytics, etc.

We will provide:

• rich data infrastructure
• verified data set

You will make core contributions to the construction for analytic tool.

Expected outcomes

• You need to implement a library NodeRank, the input is Graph(V, E) and the output is a list of V Score
• It’s better if you have optimization for execution efficiency

Skills required

• Python, Rust
• Blockchain Data
• Graph Data Mining/Analytics
• Prefer: Experience in graph databases

Mentor

Zhong Zella @ZhongFuze

Expected size of project

175h

Level

Medium

[BinaryHB0916]

Thank you for your attention. OSPP2023 claimed this idea here: https://summer-ospp.ac.cn/org/prodetail/232500141?list=org&navpage=org

Please go check out the other idea-lists. And feel free to comment on what kind of thing you would love to work on. We can arrange your favorite for your needs!