Data preprocessing and Data analysis Resume

TECHNICAL SKILLS

Programming Languages: C/C++, Scala, R, Python, Java, CSS/CSS3, HTML, Jquery, Visual basic, vbscript, JavaScript, XML
Database: MySQL, Microsoft SQL server, Ms - Access, Oracle 10g
IDE: Eclipse, Jupyuter, Rstudio
Big Data: Spark, Hadoop, MapReduce, Hive, Data Mining, Hortonworks, Machine learning, Predictive analysis, Data analysis, Data crawling
Tools: Weka, UFT (Unified Functional testing), ALM (Application Lifecycle Management)
OS: Windows, Ubuntu, CentOS

PROFESSIONAL EXPERIENCE

Confidential

Data preprocessing and Data analysis

Responsibilities:

ETL (Extract, Transform, Load) Data from Web applications, Terminal emulator (Bluezone) Mainframe, Services Like SOAP, URLs using scripting, use of excel for data manipulation
Getting requirements from clients, understanding and analyzing the business for building ETL scripts
Increased the efficiency of the scripts for Data analysis and reduced the manual work by 70 % to extract
Used UFT (Unified functional testing) vbscript and ALM (Application Lifecycle Management) for Automation and Functional testing
Strong verbal and written communication skills as a single resource for handling the module

Confidential

Software Developer

Responsibilities:

Worked on Development of Web pages in Java, JavaScript, jQuery, CSS, HTML
Created REST web services through java for webpages and IOS application using Microsoft SQL server management studio 2008 for database connectivity
Lead a team to complete CUSTARD (Cloud User Security System for Transactions through Accounting and Authentication of Resource Delivery) using Java IDE (Eclipse), Apache Tomcat, SQL.
1 + year experience of academic research using machine learning, data mining and statistical methods
Feature Engineered the most recent and previous offenses and computed the time to recidivate using statistics
Applied Logistic Regression and Random forest to achieve an accuracy of 68% for prediction of recidivism of inmates within 3 years with a false positive rate of 15% and true negative rate of 20%
Statistical data analysis and Association rule mining to determine the pattern of recidivism using Hadoop, Hive, Scala, Python, R, pandas, matplotlib, numpy, R, PySpark, Spark, Spark MLlib, Hortonworks Data platform using Machine Learning.