SONAL CHAPEGADIKAR

Detail-oriented Analyst with a proven ability to enhance data quality and operational efficiency through innovative solutions. Proficient in managing large datasets, developing GIS maps, and creating interactive dashboards. Skilled in statistical research and machine learning to ensure data integrity and reliability.

• SQL for Data Science, University of California, Davis (Coursera)
• Data Science Prodegree, Imarticus Learning
• IBM Data Science Professional Certificate, Coursera

• Detection fake job postings using text classification (NLP)

- Utilized NLP pre-processing techniques, including tokenization, lowercasing, and lemmatization, to   prepare raw textual data for machine learning algorithms.

- Transformed the data into numerical vectors using Bag of Words (BoW), TF-IDF, and word embeddings.

- Successfully deployed and tested over 5 machine learning models, achieving a classification accuracy rate of 90% for job ads.

• Use of Machine learning for fast and accurate anomaly detection in Industrial IoT

- Imported and merged network traffic data from CSV and p.cap formats into a single data frame using Pandas, processing over 1 million rows of data.

- Analyzed network packets and detected cyber attacks on Industrial IoT devices.

- Standardized models with fit-scalar functions and tuned hyperparameters, reducing Type-1 errors by 23% and improving precision by 18%.

• Prediction and Analysis Of Default Rate Future Loan

- Created a data model to forecast the likelihood of defaulters and non-defaulters with techniques like SVM, Random Forest, Bagging and Logical Regression.