I have experience working in Fintech and Banking domain.
Author of Zoofs and NitroFE
EXPERIENCE ....
✔ Data Science
Sep 2020 - Present
Experience in ML development for integrating various B2B applications, leveraging machine learning,
data regression, rule-based models, robust algorithms and techniques. Responsible for measuring and optimizing the quality of algorithms and models.
✔ Product I have worked on
-
B2B Credit Limit Suggestion
Created the product for predicting Credit Limit actions (upgrade/downgrade/extension) to be taken for B2B customers.
Historical action, Sales forecast and industry specific credit information is utilized to predict credit action to mitigate risk for the companies.
Model has been adopted by two companies to perform proactive credit management and manage defaulting customers. -
Customer Segmentation
Created the product for periodic customer segmentation of B2B customers to facilitate better collection strategies,
using clustering algorithms and a unique ELO based-customer rating system.
Also Responsible for addition of explainable k-means algorithms for cluster explanation to boost customer trust in the solution.
The final delivery helped the companies to curate collection strategies using customer segments. -
Payment date prediction
Created the product for predicting the payment date of invoices for B2B customers.
Model has been adopted by companies to proactively identify delayed invoices and act on them to improve their order to cash lifecycle.
✔ My open source work
✔ My publications
- COLLECTIVE STATE IMPLEMENTATION ON PARTICLE SWARM FOR FEATURE SELECTION
ICDICI-2020, ISBN : 978-981-15-8529-6
Binary particle search optimization has limitations of premature and slow convergence. Hence a new collective
state implementation for particle swarm optimization is proposed in this work. The proposed method is
validated with five benchmark datasets. To measure the impact of the algorithm, our proposed solution was
compared against two popular feature selection algorithms Binary Particle Swarm Optimization (BPSO) and
Genetic Algorithm (GA). Our results show that our proposed solution Collective state implementation- Particle
Swarm optimization (CSI-BPSO) achieves competitive scores in feature selection
My working tools...
