Merchant Memento
A crowdsourced database and API that deciphers cryptic merchant descriptions on bank statements, helping users and financial apps identify and understand their spending.
Temporal Payment Forensics
A predictive fraud detection system using blockchain timestamp analysis to identify anomalies suggesting 'pre-crime' financial activity, inspired by '12 Monkeys' and the cautionary themes of technological overreach in 'Frankenstein'. It focuses on micro-transactions within specific niches prone to temporal fraud.
Imagine a system, similar to the 'Army of the 12 Monkeys' attempting to understand the future to prevent a disaster. Our project, 'Temporal Payment Forensics', aims to predict fraudulent payment activities -before- they fully materialize, albeit on a smaller, niche scale. The concept stems from the inherent temporal nature of blockchain transactions, specifically their timestamps. We leverage a 'Job Listings' scraper, adapted to monitor payment platforms and cryptocurrency exchanges focused on specific niches (e.g., NFT marketplaces, freelance gig economies, crowdfunding platforms for creative projects). The scraper collects transaction data including timestamps, amounts, involved addresses/accounts, and transaction metadata.
The 'Frankenstein' influence is a cautionary tale: technology, however well-intentioned, can have unforeseen consequences. The system doesn't aim for infallible prediction but rather identifies anomalies in transaction timing. The '12 Monkeys' influence lies in proactively analyzing the past (blockchain history) to identify patterns indicating potential future fraud, specifically activities that demonstrate awareness of future events before they occur.
Here's how it works:
1. Niche Identification: Focus on payment systems with publicly accessible transaction data (even if partially obscured) and a specific vulnerability. Examples include manipulating timestamps on NFT creation to fabricate artificial scarcity, or coordinating micro-transactions in freelance markets to boost profile visibility before larger scams.
2. Data Scraping: The scraper gathers blockchain transaction data from the target niche. This involves adapting existing 'Job Listings' scraper code to parse transaction records.
3. Temporal Anomaly Detection: The core algorithm analyzes transaction timings and patterns, looking for deviations from expected behavior. This could involve:
- Transactions occurring unusually close to market events (e.g., spikes in NFT prices).
- Coordinated micro-transactions preceding larger, suspicious transfers.
- Unusual timestamp patterns inconsistent with human activity.
4. Risk Scoring: A risk score is assigned to each transaction or user based on the severity of the detected anomalies.
5. Alerting/API: The system provides an API to payment platforms. When a transaction's risk score exceeds a certain threshold, the platform receives an alert, allowing them to implement preventative measures (e.g., requiring additional verification, delaying payouts, or freezing accounts).
Low-Cost Implementation:
- Utilizes existing 'Job Listings' scraper code as a foundation, reducing development time.
- Leverages open-source libraries for data analysis and machine learning.
- Focuses on specific niches, limiting the scope of data collection and analysis.
- Cloud-based deployment keeps infrastructure costs low.
High Earning Potential:
- Payment platforms in niche markets are highly vulnerable to fraud and willing to pay for proactive detection systems.
- A successful system can be scaled to multiple niches, increasing revenue streams.
- The system can be offered as a SaaS product with tiered pricing based on transaction volume or risk score accuracy.
This project offers a unique approach to fraud detection by focusing on the temporal aspects of blockchain transactions, inspired by the proactive but ultimately flawed approaches found in '12 Monkeys' and tempered by the ethical considerations raised in 'Frankenstein'.
The 'Frankenstein' influence is a cautionary tale: technology, however well-intentioned, can have unforeseen consequences. The system doesn't aim for infallible prediction but rather identifies anomalies in transaction timing. The '12 Monkeys' influence lies in proactively analyzing the past (blockchain history) to identify patterns indicating potential future fraud, specifically activities that demonstrate awareness of future events before they occur.
Here's how it works:
1. Niche Identification: Focus on payment systems with publicly accessible transaction data (even if partially obscured) and a specific vulnerability. Examples include manipulating timestamps on NFT creation to fabricate artificial scarcity, or coordinating micro-transactions in freelance markets to boost profile visibility before larger scams.
2. Data Scraping: The scraper gathers blockchain transaction data from the target niche. This involves adapting existing 'Job Listings' scraper code to parse transaction records.
3. Temporal Anomaly Detection: The core algorithm analyzes transaction timings and patterns, looking for deviations from expected behavior. This could involve:
- Transactions occurring unusually close to market events (e.g., spikes in NFT prices).
- Coordinated micro-transactions preceding larger, suspicious transfers.
- Unusual timestamp patterns inconsistent with human activity.
4. Risk Scoring: A risk score is assigned to each transaction or user based on the severity of the detected anomalies.
5. Alerting/API: The system provides an API to payment platforms. When a transaction's risk score exceeds a certain threshold, the platform receives an alert, allowing them to implement preventative measures (e.g., requiring additional verification, delaying payouts, or freezing accounts).
Low-Cost Implementation:
- Utilizes existing 'Job Listings' scraper code as a foundation, reducing development time.
- Leverages open-source libraries for data analysis and machine learning.
- Focuses on specific niches, limiting the scope of data collection and analysis.
- Cloud-based deployment keeps infrastructure costs low.
High Earning Potential:
- Payment platforms in niche markets are highly vulnerable to fraud and willing to pay for proactive detection systems.
- A successful system can be scaled to multiple niches, increasing revenue streams.
- The system can be offered as a SaaS product with tiered pricing based on transaction volume or risk score accuracy.
This project offers a unique approach to fraud detection by focusing on the temporal aspects of blockchain transactions, inspired by the proactive but ultimately flawed approaches found in '12 Monkeys' and tempered by the ethical considerations raised in 'Frankenstein'.