Dr. Talya Eden

Algorithms and Property Testing in Big Data

When a social network attempts to identify bot networks, fake accounts, or spreaders of misinformation in real-time, it faces a complex computational challenge. Continuously scanning every user, connection, and activity pattern is impractical when dealing with millions of data points. However, time is critical: late detection of coordinated activity can allow misinformation to influence public opinion, financial markets, and public decision-making.

Dr. Talya Eden’s research focuses on developing sub-linear algorithms-methods that allow for drawing conclusions about massive datasets by examining only a tiny fraction of them. Through intelligent sampling and probabilistic calculations, these algorithms estimate the properties of large networks, identify suspicious structures, and provide reliable answers even when it is impossible to inspect all the data. Her work bridges the gap between mathematical theory and the efficient analysis of Big Data.

Primary Research Areas:

• Sub-linear Algorithms: Analyzing data in significantly less time than the total size of the dataset.

• Graph Property Testing: Identifying structural characteristics based on partial information.

• Sampling Algorithms: Estimating sizes and structures in massive datasets.

• Efficiency Analysis: Proving the correctness and accuracy of algorithms under uncertainty.

Research Nature:

 Theoretical-Applied: Developing efficient algorithms for Big Data analysis, alongside mathematical proofs for performance and accuracy.

Career Path:

 Graduates acquire advanced tools in algorithm development and analysis, sampling, and mathematical thinking, transitioning into various paths:

• Industry: Data Science, algorithm design, and network analysis.

• Big Data Systems: Developing efficient methods for analyzing and managing massive-scale information.

• Academia: Research in algorithms and theoretical computer science.