Research Themes | Rinku Jacob

Research focus

My research focuses on extracting meaningful structure from complex data. A time series, signal, light curve, trajectory or image-derived dataset can be converted into a mathematical representation such as a phase-space reconstruction, recurrence plot, network or topological descriptor.

The purpose is not only classification. The main aim is interpretation: to understand how the system behaves, changes, organises itself, or moves from one dynamical state to another.

Central method: observed data → reconstructed representation → measurable structure → physical or scientific interpretation.

Core research themes

Each theme below represents a real direction in my research, publications, or current academic development.

∿

Nonlinear time series analysis

Study of signals from nonlinear and chaotic systems using delay embedding, phase-space reconstruction and recurrence-based analysis.

Chaos Embedding Signals

◉

Recurrence plots and recurrence networks

Conversion of repeated states in phase space into recurrence plots and networks to study geometry, structure and changes in dynamical systems.

RP RN Phase space

⛓

Complex network measures

Use of graph measures such as link density, clustering, path length and degree heterogeneity to quantify organisation in reconstructed data.

Networks Link density Heterogeneity

↯

Dynamical transition detection

Tracking changes in nonlinear systems using recurrence-network features, especially link density and sliding-window analysis.

Transitions Windows Change points

♥

Biomedical signal analysis

Application of recurrence-network topology and amplitude-preserving signal representations for ECG analysis and physiologically interpretable classification.

ECG Cardiac dynamics Classification

△

Topological data analysis

Use of topological and geometric descriptors for complex datasets such as trajectories, image-derived point clouds and evolving structures.

Topology Persistence Point clouds

✦

Astrophysical and radio-astronomy data

Extension of nonlinear, recurrence and network-based thinking to astronomical variability, light curves and radio-astronomy data-analysis workflows.

Light curves MeerKAT Radio data

⚙

Scientific computing workflows

Development and use of computational workflows in Python, MATLAB and Jupyter for signal analysis, network measures, visualisation and reproducible research.

Python MATLAB Jupyter

AI-native science education

Academic work on how artificial intelligence changes scientific learning, physics pedagogy, teacher training and student research formation.

AI-native Physics learning Pedagogy

Real outputs connected to these themes

These are examples of actual work connected with the themes above.

Doctoral research Ph.D. thesis on nonlinear time series analysis using complex network measures.

Recurrence-network publications Work on recurrence networks, weighted recurrence networks, degree heterogeneity and recurrence-network measures.

Biomedical signal publication ECG classification using recurrence-network topology and amplitude-preserving normalisation.

Topological methods publications Topological approaches applied to trajectory analysis and CT scan point-cloud based classification.

Radio-astronomy development Current academic direction connected with MeerKAT data-analysis training, CASA, CARTA and astronomy workflows.

AI-native education work Workshop and curriculum-related work on AI-native physics learning and teacher training.

The common aim across these themes is to make complex data scientifically interpretable, not merely computationally processed.

For students and collaborators

Students can begin with small, well-defined problems in signal analysis, recurrence plots, networks, ECG data, astronomy data, topology or scientific coding. Collaborators may contact me for interdisciplinary datasets where nonlinear, network or topological methods can add interpretation.

Student project areas Contact for collaboration View publications