Abhimanyu Saraf Email & Phone Number

Seasoned data science practitioner with rich multi-disciplinary experience of marketing research, business analytics, data sciences and product development, and coverage of wide gamut of industries (consumer electronics, CPG, online media, pharmaceuticals, telecom, financial services, hospitality etc.). Presently, playing the role of subject matter expert (data sciences) in the products and innovation labs. Have had hands-on experience on open-source tools like R, Python etc., data technologies (for storage and analysis of large amounts of data) like AWS S3 and others like SPSS, VBA & MS-Office suite for data pre-processing, data preparation, data mining, data visualization, predictive modeling, model evaluation & validation, and interpretation of outputs.Proficiency in executing Machine Learning techniques like:— Regression (Linear, Ridge, LASSO, Elastic Net, LARS, Stochastic Gradient Descent, Bayesian, Robust, Isotonic, Cubist, Correlated Component etc.)— Classification [Logistic Regression, Ordinal Logistic Regression, k-Nearest Neighbour, Naïve Bayes, Discriminant Analysis (LDA, QDA), Survival Analysis (Kaplan-Meier Estimator, Cox’s Proportional Hazards Model), Hidden Markov Model, Stochastic Gradient Descent]— Ensemble Methods (Bagging, Random Forest, Gradient Boosting Machine, AdaBoost, LightGBM, XGBoost, Ensemble Segmentation etc.)— Decision Trees (CART, CHAID, QUEST, ID3, C4.5, C5.0, Conditional Decision Trees etc.) — Support Vector Machines (Regression, Classification, Clustering, Anomaly Detection, Density Estimation)— Pathway Analysis (Structural Equation Modelling, Partial Least Squares, Bayesian Belief Network)— Clustering [Non-hierarchical (k-Means, k-Medoids, k-Modes, k-Prototype), Two-step (BIRCH), Hierarchical, Expectation Maximization, Spectral, Latent Class, Hidden Markov Model etc.]— Dimension Reduction (Principal Component Analysis, Multidimensional Scaling etc.)— Recommender Systems (Collaborative Filtering, Matrix Factorization)— Meta-learning Methods (Stacked Generalization, Cascade Hybrid, Feature Augmentation etc.) — Reinforcement Learning (Markov Decision Process, Dynamic Programming, Monte Carlo methods, Q-Learning, Temporal Difference Learning)— Deep Learning (Artificial Neural Network)— Resampling Methods (SMOTE, Case-control Sampling, Bootstrapping, Cross-validation)— Other techniques like Conjoint Analysis (Discrete Choice Modelling), Marketing Mix Modelling, Total Unduplicated Reach Frequency (TURF), Price Sentivity Meter, Penalty Reward Contrast Analysis, Key Driver Analysis etc.

Listed skills include Analytics, Market Research, Data Analysis, Sas, and 11 others.