Prateek Sharma Email & Phone Number

Gurgaon, India

Gurgaon, India

Gurgaon

Client : Ultimate automotive marketplace for car dealers, manufacturers and individuals to advertise and market their vehicles to millions of in-market car shoppers.Provides potential buyers with reviews, shopping advice, and comparison tools for car financing and insurance informationProject: To deep dive in consumer analytics on the website and to derive useful insights. Thereby, helping the business in strategy decision making.-- Tools used are Adobe Analytics, Microstrategy web, Netezza(SQL database), Excel and Powerpoint.

Gurgaon

Product Development Project- Built an end to end Sentiment Analysis tool --Researched and devised a Hybrid algorithm using supervised and unsupervised learning techniques-- This Algorithm branches out and produce results at different granularity depending on the business problem to be addressed.i) Document Level Analysis - We extract subjective(opinionated) sentences by either of the two techniques, namely, Bootstrapping and Lexicon and Adjective based classification. The subjective sentences are then classified into different categories of polarity(positive, negative or neutral) using either of the three machine learning techniques, namely, Support Vector Machine, Naïve Bayes and Expectation Maximization (Using Naïve Bayes).If the business wants to know the overall sentiment, “Are customers happy or not?”, then document level analysis would sufficeii) Aspect level analysis- Aspect signify different features of a product/service. This analysis involves aspects extraction and use rule and lexicon based methods for aspect level sentiment rating. If the business wants to find the aspects of the product which make customers happy or to know the areas of improvement, then aspect level analysis would serve the purpose.-- Closely worked with teams to successfully implement the algorithm in R and Python--Fetched tweets about a global hamburger fast food restaurant chain spread across 119 countries. A manually created sampling algorithm was used to select 1397 of these tweets.These were fed into our algorithm, We obtained an accuracy of 72 % .--Co-wrote “Sentiment Analysis Triology”, a series of three articles,which focuses on exploring and unraveling Sentiment analysis considering all its business implicationsFind the links to the “Sentiment Analysis Triology” articles below .

Gurgaon, India

Client- Leading American pharmaceuticals distributor at a retail sale level providing medical supplies and care management tools.Project- To Prepare a Market Share Reporting document to provide insights into client’s position versus the market and to highlight potential improvement opportunities.-- Document detailed on Client’s market share, growth across its business segments and regional details.-- Assessed the business rule changes and its impact on the structure of the existing template used for generation of the document.-- Figured out inconsistency in the data by effective methods of Quality checks.-- Rectified a method of aggregating sales resulting in a correction of $14.7 mn.-- A more granular approach to the project in synergy with Business rules rectification resulted in an increased revenue of 150%.Client- Esteemed American multinational, spread across 40 countries, providing commercial & consulting services to multinational Pharmaceutical industriesProject- To build a set of insightful and dynamic dashboards, handling large amounts of data to help identify and develop new strategic business opportunities for pharmaceutical firms.--This new system enabled over a 100 reps/managers to analyse their sales on a daily basis and strategize to attain their goals.--Supported three pharmaceutical giants by analysing the data discrepancy problems; debugging and deciphering the corresponding issues both in microstrategy environment and in the data warehouse .Reduced the time taken to solve issues by an average of 20%.--Acted as a direct point of contact for medical reps, managers and onshore team for data related problems.--Acted as a bridging point between the Business intelligence and Data warehouse teams.--Presided over meetings and discussions, for briefing the issues faced by the client .Thereby, leading to effective solutions.

Mumbai Area, India

Project: Characterization of machined samples of SS-304L and Zircaloy-4 Performed micro-structural characterization of machined samples: SS304L and Zircaloy 4 using various techniques (optical-microscopy, FEG-SEM, TEM, Glancing Angle XRD and EBSD) to study the effect of surface finishing and machining in practical applications. Detected surface material phase transformation proving detrimental effects of machining on these materials.

Pgdm

Bachelor'S Degree, Materials Engineering, 7.97/10

Activities and Societies: General Secretary – Alumni Relations Cell(ARC), IIT Roorkee (Aug’12 – May’14) Vice President – Metallurgical.

X Cbse Board- 93%, Xii Cbse Board- 91.2%