Pavithra Kumar Email & Phone Number

Charlotte, Nc, Us

Espoo, Fi

Serving as the primary Data Scientist at Ahlstrom US, I lead the charge in applying machine learning to transform our paper pulp recipes. My efforts are centered on process optimization and innovation, ensuring high-quality outcomes. I partner with plant managers, engineers, and R&D teams to push the boundaries of our industry. I support company vision on Sustainability for a better world with optimizing fibers using advanced AI techniques.

New York, Us

New York, Ny, Us

• Evaluated models using Cross Validation, Log loss function, ROC curves and used AUC for feature selection and elastic technologies like ElasticSearch, Kibana etc• Addressed overfitting by implementing of the algorithm regularization methods like L2 and L1. • Implemented statistical modeling with XGBoost machine learning software package using Python to determine the predicted probabilities of each model.• Created master data for modelling by combining various tables and derived fields from client data and students LORs, essays and various performance metrics.• Hands on Experience on Spark MLLib using both Python and Scala• Used numpy, scipy, pandas, nltk(Natural Language Processing Toolkit),matplotlib to build the model.• Application of various Artificial Intelligence(AI)/machine learning algorithms and statistical modeling like decision trees,text analytics, natural language processing(NLP), supervised and unsupervised, regression models.• Used Principal Component Analysis in feature engineering to analyze high dimensional data. • Performed Data Cleaning, features scaling, features engineering using pandas and numpy packages in python and build models using deep learning frameworks.• Scheduled jobs to run SQL statements to transfer data from Google Bigquery to Mysql using Airflow

Beaumont, Tx, Us

• Performed Data Profiling to learn about behavior with various features such as traffic pattern, location, time, Date and Time etc. • Application of various Artificial Intelligence(AI)/machine learning algorithms and statistical modeling like decision trees,text analytics, natural language processing(NLP), supervised and unsupervised, regression models, social network analysis, neural networks, deep learning, SVM, clustering to identify Volume using scikit-learn package in python, Matlab• Categorised comments into positive and negative clusters from different social networking sites using Sentiment Analysis and Text Analytics• Analyze traffic patterns by calculating autocorrelation with different time lags. • Sets of comment streams are processed by NLP module in stages like stemming, punctuation removal etc. to categorize into vector model and input into Incremental Short Text Summarization (ISTS) algorithm. Clusters of comments with same pattern will be grouped together and presented visually to the end user.• Ensured that the model has low False Positive Rate and Text classification and sentiment analysis for unstructured and semi-structured data. • Addressed overfitting by implementing of the algorithm regularization methods like L2 and L1. • Used Principal Component Analysis in feature engineering to analyze high dimensional data. • Performed Data Cleaning, features scaling, features engineering using pandas and numpy packages in python and build models using SAP Predictive Analytics• Created Data Quality Scripts using SQL and Hive to validate successful data load and quality of the data. Created various types of data visualizations using Python and Tableau

Dublin 2, Ie

• Used R and python for Exploratory Data Analysis, A/B testing, HQL, VQL, Data Lake, AWS Redshift, oozie, pySpark, Anova test and Hypothesis test to compare and identify the effectiveness of Creative Campaigns. • Computing A/B testing frameworks, clickstream and time spent databases using Airflow• Created clusters to Control and test groups and conducted group campaigns using Text Analytics. • Created positive and negative clusters from merchant’s transaction using Sentiment Analysis to test the authenticity of transactions and resolve any chargebacks.• Used Python, R, SQL to create Statistical algorithms involving Multivariate Regression, Linear Regression, Logistic Regression, PCA, Random forest models, Decision trees, Support Vector Machine for estimating the risks of welfare dependency. • Identified and targeted welfare high-risk groups with Machine learning/deep learning algorithms. • Conducted campaigns and run real-time trials to determine what works fast and track the impact of different initiatives.• Developed Tableau visualizations and dashboards using Tableau Desktop

Master'S Degree, Computer Science

Bachelor Of Technology - Btech, Computer Science