Pratik Abnave
About Pratik Abnave
Pratik Abnave is a Data Scientist I at Turing.com, recognized for developing a recommendation system that achieved 92% accuracy. He holds a Master of Computer Applications and has experience in various data science projects, including fraud detection and customer churn prediction.
Work at Turing
Pratik Abnave has been employed at Turing.com as a Data Scientist I since 2022. In this role, he developed a recommendation system utilizing deep learning techniques, which achieved a 92% accuracy rate. This development enhanced product marketing collaboration and increased campaign success by 7%. His work at Turing is conducted remotely, allowing him to contribute effectively to the organization from any location.
Education and Expertise
Pratik Abnave holds a Master of Computer Applications (MCA) degree from LATE BHAUSAHEB HIRAY SMARNIKA SAMITI TRUSTS COLLEGE OF ARCHITECTURE, where he studied from 2018 to 2021. Prior to this, he earned a Bachelor of Science in Information Technology from Nirmala Memorial Foundation Degree College of Commerce and Science between 2015 and 2018. He also possesses several certifications, including a Data Science Certification from The Interface, a Google Data Analytics Professional Certificate from Coursera, and an IBM Data Science Professional Certificate from Coursera.
Professional Background
Before joining Turing, Pratik worked as a Data Scientist at Neoperk Technologies Private Limited from 2021 to 2022. He contributed to a nutrient and soil-based crop recommendation system, achieving up to 98% accuracy. Additionally, he served as a Software Developer Trainee at EddyTools Tech Solutions for four months in 2021. Currently, he is also working at Sylvr as a Data Scientist since 2023.
Technical Projects and Contributions
Pratik Abnave has developed several notable projects in the field of data science. He created a Multi-Cure Health WebApp using Flask, capable of detecting nearly 10 diseases with an accuracy range of 77%-99%. He also implemented a credit card fraud detection project that achieved 99% accuracy. His work includes developing a Deep Learning LSTM model for sequence-based text datasets, which reached 90% accuracy and reduced manual effort by 80%. Additionally, he has implemented a customer churn prediction WebApp integrated with GitHub CI/CD pipelines and Docker.
Achievements in Machine Learning
Pratik has demonstrated significant expertise in machine learning through various projects. He developed a feature extraction function using chroma and MFCC to optimize classification accuracy for the MLPClassifier, which resulted in a 40% reduction in misclassification. His contributions to the field reflect a strong understanding of data science principles and practical applications.