I'm Priyesh Dave

About Me

I am Priyesh Dave, Data science professional with 3.8 years of experience around Machine Learning, Natural Language Processing, Deep Learning, Predictive Modeling, and Statistical Analysis.

Passionate about leveraging data science and ML to build products & services which help solve real-world business problems. Motivated to learn, grow and excel in Artificial Intelligence Industry. Presently working as Data Scientist at Tata Consultancy Services in a small agile team for building machine learning models for Westpac Bank and customized core banking software suite.

Adept at organizing and executing complex projects that deliver exceptional results across an array of endeavors in both the profit and non-profit sectors. Presently seeking for Data Scientist and Machine Learning role with long-term goal of being meaningful contributor to an organization.

Open source Outreach and working team member at Omdena, a collaborative platform to build innovative, ethical, and efficient AI and Data Science solutions to real-world problems.

Experience

Tata Consultancy Services

Working with exceptional Westpac team to build cutting-edge ML/AI solutions of banking suits. Responsible for building Machine Learning model for payments systems that identifies the customers with abusive comments in payments systems.

• Redesigned Westpac’s data masking app by using NLP which masks the confidential user information from the production logs.

• Implemented FraudPay model to detect the fraudulent payments search used by Westpac’s WDP environment.

• Working with exceptional Westpac team to build cutting-edge ML/AI solutions of banking suits.

• Responsible for building Machine Learning model for payments systems that identifies the customers with abusive comments in payments systems.

• Redesigned Westpac’s data masking app by using NLP which masks the confidential user information from the production logs.

• Implemented FraudPay model to detect the fraudulent payments search used by Westpac’s WDP environment.

Omdena

Open source Outreach and working team member at Omdena, a collaborative platform to build innovative, ethical, and efficient AI and Data Science solutions to real-world problems.

Project 1: Document Redaction

Project 2: Sustainability Benchmarking System

Projects

I have working on various real world projects ranging from classic Machine Learning to Deep Learning including applications for Natural Language Processing and Computer Vision.

Project 1: Document Redaction

This project revolves around the ability to recognize sensitive words within documents. To do this I am making use of Natural Language Processing(NLP) where the focus is on Named Entity Recognition which searches a body of text and classifies named entities into predefined categories.

• GitHub

Project 2: Intent Matcher Application

Developed a hybrid model to predict intent of two questions based on generating a context-rich text feature vector by combining single words and word sequences to form a multi-dimensional word embedding structure using Word2Vec model.

• GitHub

Project 3: Auto Text Generation

Developed an intelligent application that predicts the next n words given a sequence of words based on a Markov's N-Grams assumption.

• GitHub

Project 4: Movies Recommender System

Developed an intelligent application that recommends user movies related to the movie selected by the them based on Cosine Similarity among the movies.

• GitHub

Project 5: Road Traffic Severity Predictor

RTA severity predictor is an application which predicts the severity of road traffic accident, so as to pave the way for improving the safety level of road traffic.

• GitHub

Project 6: Patient Survival Predictor

Developed and verified a Deep Learning model for predicting the risk of death in ICU patients. Because the traditional regression model predicts mortality rate based on a simplified relationship between predictor variables and prognosis, it is difficult to improve the predictive ability of that model. Therefore, based on the strong nonlinear fitting ability and more precise algorithms of deep learning, I constructed a neural-network model for predicting the survival rate of patients in the ICU's and explored the influencing patient factors.

• GitHub

Project 7: Cyberbully Detector

Developed and verified a Deep Learning (BidirectionalLSTM) model that classifies whether a given comment is toxic or not.

• GitHub

Applied ML

Implemented Machine Learning algorithms from scratch using python.

• GitHub

Certifications

AWS INNOVATE AI & ML EDITION

AWS Cloud Practitioner Essentials

Accenture AI Digital

Machine Learning Engineer

The Machine Learning Company

Ultimate AWS Cloud Practitioner

UMICH - Intermediate PostgreSQL

UMICH - Basic PostgreSQL

Contact

• LinkedIn
• GitHub
• Twitter

Accomplishments

• Icon

Elements

i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';