SUMMARY:

• Multiple years of IT experience and Analytics experience in data analysis, strategic analysis, Predictive Analytics and Inferential statistics, data mining and data visualization to provide analytical solutions to client's business problems.
• Strong foundation in Mathematics, Statistics, Machine Learning and Data Mining techniques.
• Proficient in managing entire data science project life cycle and actively involved in all the phases of project life cycle including
• Data acquisition (sampling methods: SRS/stratified/cluster/systematic/multistage)
• Power Analysis, Hypothesis testing, effect size
• EDA (Univariate & Multivariate analysis)
• Data cleaning
• Data Imputation (outlier detection via chi sq detection, residual analysis, PCA analysis, multivariate outlier detection)
• Data Transformation
• Features scaling
• Features engineering
• Statistical modeling both linear and nonlinear (logistic, linear, Naïve Bayes, decision trees, Random forest, neural networks, SVM, clustering, KNN)
• Dimensionality reduction using Principal Component Analysis (PCA) and Factor Analysis, testing and validation using ROC plot, K - fold cross validation, statistical significance testing.
• Data visualization.
• Worked on Reinforcement Learning Techniques like Multi Armed Bandit (UCB & Thompson Sampling) A/B testing.
• Experience in Data Analysis obtain insights from data then choose appropriate Machine Learning/Data Mining algorithms (Functional/classification/cluster models etc) using cost functions (gradient/normal and stochastic)
• Perform model validation (AIC/BIC/APE/MAPE, confusion matrix, ROC curve, KS curve, decile analysis, Gini coefficient)
• Experience with various application like SAS, SPSS, R(packages- knitr,dplyr, tidyr, SparkR, causallnfer, spacetime), Python (sklearn/ scipy/numpy/panda).

PROFESSIONAL EXPERIENCE:

Confidential, Jersey City, NJ

Data Scientist

Responsibilities:

• Providing consultation services and technical development for a Fortune-20 BFSI client on new customer acquisition, database enrichment, quality improvement and decision making.
• Personalization, Inflection points tracking, Target Marketing, Customer profiling and Segmentation
• Worked on model building using machine learning algorithms like logistic regression, naïve bayes, random forest, SVM, SVR.
• Perform analysis on Apache Hadoop (Hive) and generate reports on data quality including data accuracy.
• Developed pipeline using Hive (HQL) to retrieve the data from Hadoop cluster, SQL to retrieve data from Oracle database and used ETL for data transformation.
• Used Python based data manipulation and visualization tools such as Pandas, Matplotlib, Seaborn to clean corrupted data before generating business requested reports
• Used Python, R and Spark to develop variety of models and algorithms for analytic purposes.
• Developed NLP service to identify and extract text features for pre-populating fields in the client's data reporting and abstraction application.
• Profiling based on the content access reports, event trigger reports and navigation reports. Profiles are then used for target marketing and segmentation
• Quantify the incremental value of data obtained from various data sources and perform arbitration of the data using model
• Worked on a Proprietary tool to reconcile the vendor database with that of client’s information
• Well experienced in Normalization & De-Normalization techniques for optimum performance in relational and dimensional database environments.
• Data analysis using regressions, data cleaning, excel v-look up, histograms and TOAD client and data representation of the analysis and suggested solutions for investors.
• Experience in Artificial Neural Networks, Deep Learning models using Theano, Tensorflow and keras packages using Python.

Confidential, Tampa, FL

Data Scientist

Responsibilities:

• Personalization for registered users, Returning visitors and Anonymous visitors
• Worked on a POC project for NLP (Natural Language processing).
• Propensity to convert modelling using the visit sequence and conversion sequence mapping
• Zeroth problem solution for Anonymous visitor, returning visitor and registered users
• Text analytics on review data machine learning technique in python using NLTK.
• Market basket Analysis using APRIOR for purchase transactions data at store level to recommend customers with the different combinations of the products. Custom Dashboards product category wise.
• Sequence mining for the conversion paths for different segments
• Site/Page optimization, Promotional campaigns assessment, Customer segmentation and assessing revenue against targets at regular intervals
• Used UCB & Thomson Sampling Intuition for Multi Bandit Testing and A/B testing to perform ad banner choices and product visual choices optimization.
• Developed visualizations using sets, Parameters, Calculated Fields, Dynamic sorting, Filtering, Parameter driven analysis, gathered data from different data marts.
• Applied advanced machine learning algorithms including PCA, K-nearest neighbors , random forest, gradient boosting, neural network and xgboost to predict weights and labels of Higgs Boson with high accuracy
• Reporting designs based on the business specific problems, Reporting implementation on Tableau .
• Advanced charts, drill downs and intractability are incorporated in the reporting for different stakeholders and Integrating the publishing of reports to the clients SharePoint infrastructure

Confidential, Jacksonville, FL

Data Scientist

Responsibilities:

• Participated in all phases of data mining- data collection, data cleaning, developing models, validation, visualization and performed Gap analysis.
• Providing Ad hoc analysis and reports to Executive level management team.
• Captured Modelling requirements from Senior Stakeholders to Bench functional requirements for SAS/ R Python
• Performed Data Manipulation and Aggregation from Various source including HDFS.
• Creating various B2B Predictive and descriptive analytics using R and Tableau
• Used pandas, numpy , seaborn, scipy, matplotlib, scikit-learn in Python for developing various machine learning algorithms.
• Designed and tested Predictive Algorithms using Historical Data
• Utilized machine learning algorithms such as Decision Tree, linear regression, multivariate regression, Naive Bayes, Random Forests, K-means, & KNN.
• Parsing data, producing concise conclusions from raw data in a clean, well-structured and easily maintainable format.
• Responsible for Big data initiatives and engagement including analysis, brainstorming, POC, and architecture.
• Worked on different data formats such as JSON, XML and performed machine learning algorithms in R
• Hands on experience in implementing LDA, Naive Bayes and skilled in Decision Trees, Random Forests, Linear and Logistic Regression, SVM, Clustering, neural networks working along with operations team to move non secured cluster to secured cluster
• Worked on Map Reduce/Spark Python modules for machine learning & predictive analytics in Hadoop on AWS .
• Worked with (Tableau) Report Writers to Test, Validate Data Integrity of Reports

Confidential, Peoria, IL

Modeling Analyst

Responsibilities:

• Prepared financials reports, analysis of month on month variance and provided reports to management on a weekly, monthly and quarterly basis
• Scrutiny of General Ledger on monthly basis and do a financial analysis of the same. Provide the information in an appropriate manner to the management
• Monitored costs & trend analysis and providing updates on variances to respective stakeholders
• Prepared annual budget and periodic forecast/outlook and Generating Insights
• Partnered with Leadership to understand the requirement and provide reports, Analyze business units
• Recommended solutions for areas of concern with financial data.
• Provided ad-hoc reports as per requirement from time to time