# aditya-ailsinghani/multimodal-fraud-detection architecture

> Architecture diagram of the GitHub repository aditya-ailsinghani/multimodal-fraud-detection, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-09-26). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/aditya-ailsinghani/multimodal-fraud-detection
- Repository: https://github.com/aditya-ailsinghani/multimodal-fraud-detection

## Overview

This repository is a notebook-based multimodal fraud detection project. Its workflow loads and explores IEEE-CIS transaction data, builds tabular, text-derived, and transaction-network signals in parallel, then combines them in a LightGBM fusion model for a final fraud score and feature-importance analysis. The README documents these relationships; no source excerpts were available to verify finer-grained wiring.

Main entry points: [`notebooks/05_fusion_explainability.ipynb`](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/05_fusion_explainability.ipynb) (Fusion model), [`notebooks/01_data_loading_eda.ipynb`](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/01_data_loading_eda.ipynb) (Load and explore), [`notebooks/02_tabular_model.ipynb`](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/02_tabular_model.ipynb) (Tabular model), [`notebooks/03_nlp_module.ipynb`](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/03_nlp_module.ipynb) (NLP features)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Data preparation"]
  node_loading["Load and explore"]
end

subgraph group_signals["Detection signals"]
  node_tabular["Tabular model"]
  node_nlp["NLP features"]
  node_graph["Graph features"]
end

subgraph group_fusion["Fusion and analysis"]
  node_fusion["Fusion model"]
  node_explain["Feature analysis"]
end

node_dataset[("IEEE-CIS data")]
node_fraudscore["Fraud assessment"]

node_dataset -->|"loads"| node_loading
node_loading -->|"provides data"| node_tabular
node_loading -->|"provides data"| node_nlp
node_loading -->|"provides data"| node_graph
node_tabular -->|"provides score"| node_fusion
node_nlp -->|"provides features"| node_fusion
node_graph -->|"provides features"| node_fusion
node_fusion -->|"produces score"| node_fraudscore
node_fusion -->|"analyzes importance"| node_explain

click node_loading "https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/01_data_loading_eda.ipynb"
click node_tabular "https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/02_tabular_model.ipynb"
click node_nlp "https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/03_nlp_module.ipynb"
click node_graph "https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/04_graph_module.ipynb"
click node_fusion "https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/05_fusion_explainability.ipynb"
click node_explain "https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/05_fusion_explainability.ipynb"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_loading toneBlue
class node_tabular,node_nlp,node_graph,node_dataset toneAmber
class node_fusion,node_explain toneMint
class node_fraudscore toneIndigo
```

## Components

### Data preparation

- [Load and explore](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/01_data_loading_eda.ipynb): `notebooks/01_data_loading_eda.ipynb`

### Detection signals

- [Tabular model](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/02_tabular_model.ipynb): `notebooks/02_tabular_model.ipynb`
- [NLP features](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/03_nlp_module.ipynb): `notebooks/03_nlp_module.ipynb`
- [Graph features](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/04_graph_module.ipynb): `notebooks/04_graph_module.ipynb`

### Fusion and analysis

- [Fusion model](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/05_fusion_explainability.ipynb): `notebooks/05_fusion_explainability.ipynb`
- [Feature analysis](https://github.com/aditya-ailsinghani/multimodal-fraud-detection/blob/main/notebooks/05_fusion_explainability.ipynb): `notebooks/05_fusion_explainability.ipynb`

### Other components

- **IEEE-CIS data**
- **Fraud assessment**

## Connections

- IEEE-CIS data → Load and explore: loads
- Load and explore → Tabular model: provides data
- Load and explore → NLP features: provides data
- Load and explore → Graph features: provides data
- Tabular model → Fusion model: provides score
- NLP features → Fusion model: provides features
- Graph features → Fusion model: provides features
- Fusion model → Fraud assessment: produces score
- Fusion model → Feature analysis: analyzes importance