A Multiple Salient Features-Based User Identification across Social Media.

Area of Science:

• Computer Science
• Social Network Analysis
• Information Retrieval

Background:

• Cross-platform user identification is crucial for applications like user behavior prediction and recommendation systems.
• Existing methods often rely on single attributes, limiting their effectiveness.
• Integrating diverse user data can enhance identification accuracy.

Purpose of the Study:

• To propose a novel method for cross-platform user identification using multiple salient features (MSF-UI).
• To enhance the accuracy and universality of user identification across social media.
• To compare the proposed method against existing baseline algorithms.

Main Methods:

• Extracting and fusing redundant features from user display names, network topology, and published content.
• Utilizing a multi-module calculation to determine similarity between different user features.
• Applying a bidirectional stable marriage matching algorithm for final user identification.

Main Results:

• Integrating multi-dimensional user information optimizes the universality of user identification compared to single-attribute methods.
• The MSF-UI method significantly improves precision, recall, and F1 scores over baseline methods like RCM and RFCA-SMM.
• The proposed approach demonstrates superior performance in cross-platform user identification.

Conclusions:

• The MSF-UI method effectively identifies users across social media by leveraging diverse features.
• This approach offers a more robust and accurate solution for cross-platform user identification.
• The findings have implications for user behavior analysis, recommendation systems, and information retrieval.