Data Scientist Resume

PROFESSIONAL SUMMARY:

6.5 years of experience in Machine Learning, Predictive Analytics, Data Mining and ETL Development
Had architected solutions for problems in Machine Learning, Advanced Predictive Analytics, Business Optimization and Text Mining
Strong understanding of Big Data concepts: HDFS, Map - Reduce and Hadoop Eco- system
Good knowledge of SQL/PLSQL
Knowledge in Object Oriented Programming concepts
Proven ability as a quick learner of new skills and technologies
Effective team player with excellent communication skills and an insight to determine priorities, schedule work and meet critical deadlines
Ability to rapidly troubleshoot and resolve complex technical issues
Strong analytical, problem solving, programming and debugging skills

TECHNICAL SKILLS

Languages: R, Python, Core Java, C, PL/SQL

Platform/Technologies: HDFS, Map-Reduce, Hive, HBase, Mahout, nltk

Tools: Talend 4.1.2, Quick Rec, Tortoise SVN, Putty, Cloudera Manager

Essential Skills: Text Mining, Data Mining, Data Modeling, Predictive Analytics, Statistical Modeling, Big Data, Advanced Analytics Algorithms

PROFESSIONAL EXPERIENCE

Confidential

Data Scientist

Responsibilities:

Collection of Data, Dividing the data into Training, Testing and Evaluation
Designing, Architecting and Choice of Algorithms
Model Building using Random Forest and Gradient Boosting methods like the XGBoost
Feature Selection from the fitted model
Evaluation of the Model built
Visualization of the Output
Analyzing the False Negatives, False Positives

Environment: R

Analytics Engineer

Confidential

Responsibilities

Environment - Revo R, DeployR

Confidential

Analytics Engineer

Responsibilities:

Environment - R, Java, Revo R, DeployR

Confidential

Analytics Engineer

Responsibilities:

Text Classification

Confidential

Responsibilities:

First being a probability based approach, The NaïveBayes Classifier, which uses Bayes Theorem to predict the probability that a given feature set belongs to a particular label.
Second being a linguistic based knowledge incorporation model for identification of context and then incorporating Nearest-Neighbor Classifier for text classification. The model represents text in terms of synsets in the WordNet- a lexical knowledge base of English words along with the semantic relations. WordNet similarity is measured between the word(from the input, the tweet) and words from a manually prepared medical dictionary. Having captured the relatedness of a tweet to a medical term in the med dictionary, the data is set, to run the K- Nearest-Neighbor Classifier.:
Collection of Data, Dividing the data into Training, Testing and Evaluation
Manual labeling of the class for Training and Testing
Designing, Architecting and Choice of Algorithms
Model Building in Python
Evaluation of the Model built
Visualization of the Output in Python
Analyzing the False Negatives, False Positives

Confidential

ETL Developer

Roles & Responsibilities:

Offshore lead role for E2E AML RECON
Involved in requirement gathering and analysis of the raw data
Involved in Development of the ETL job in Talend for application of rules on raw transaction files for various Product Processors
Involved in Unit testing of the ETL jobs created
Involved in preparing Unit Test Case documents for each of the ETL jobs created
Involved in running the QuickRec tool to verify the differences in the Expected and the Actual results
Involved in running the QuickRec tool to verify the differences in the Expected and the Actual results
Involved in Break Analysis of the output to determine whether the correct rules have been applied
Interact with the clients on Break analysis

Technology - Core Java, SQL, UNIX, Talend 4.1.2, Quick Rec, Tortoise SVN, Putty