Rohit Gavval

Senior data scientist with a rich experience of analyzing and building models for structured and unstructured data. Involved in multiple end-to-end data science projects and possess hands-on experience in all phases of the data science life cycle.

Visualization of bank customer complaints with CUDASOM

• Developed a visual analytics framework for product-wise visualization of customer complaints using unsupervised neural networks
• Built an enhanced version of the tool using NVIDIA CUDA

Segmentation of customer complaints to expedite grievance resolution

• Designed a novel multiobjective clustering algorithm using swarm intelligence techniques to achieve product-wise segmentation of bank related customer complaints extracted from online complaints forum
• Applied rigorous text mining procedures for processing raw data obtained from the website, in addition to annotation of the data. Worked with several text representations and reported the best representation for the task

Credit scoring model for German credit data

• Performed extensive EDA using R and built several models using R and Weka
• Performed statistical tests and compared the models built to propose the best model

Visual sentiment analysis of bank customer complaints using parallel self-organizing maps. Handbook of Big Data Analytics: Applications in ICT, Security and Business Analytics, Volume 2

Enterprise Class Deep Neural Network Architecture for recognizing objects and faces for surveillance systems. COMSNETS 2019

Sentiment Classification of Indian Banks' Customer Complaints. TENCON 2019