Project Web Services

which is written using the sentiment API in R language.

● The Analysis Engine Streams the text files from Hadoop and performs sentiment analysis on them and provides the scores for polarity and emotions.

● The analysis algorithm is built by training the naïve Bayes algorithm trained on subjective lexicon and simple voter algorithm.

Environement: H adoop, Java, Python, pocketsphinx, R, Jboss app server. Project Name:R eport Generation on Sales Data.

Client:

Task 1:G enerating Reports for front counter transactions and gateway counter transactions

.

Task 2: C onfiguring the entire workflow for the below process using Apache oozie. The above workflow is configured in Oozie which is scheduled to run once in a week. Task 3: Coding Mapreduce for the following use case: Challenge:F iles present on s3 were copied to ec2 instances and these archive files were extracted on local file system of ec2

and then were copied to HDFS from local file system which was approximately taking 8hrs of time. Solution:C oded a Map Reduce Job which Extracted archive files from s3 to hdfs directly in a distributed fashion reducing the extraction time to 2hrs.

Environment:J ava,Hadoop,Hive,Pig,Oozie.Amazon Web Services(Ec2,s3). Project Name : e Cat

Client:

Description: Enterprise Metadata Management Tool. Role:H adoop Engineering

Task 1: Creating delta reports by connecting to hive metastore and differentiating the tables present in eCat and Hadoop Cluster.

Task 2: Creating a report which will fetch the last 20 accessed objects on the hadoop cluster. Task 3: Creating a report which will alert whenever the metadata of the tables are changed in hadoop cluster. Task 4:D ata Modelling of the project.

Task 5: Reverse Engineering of already existing tables in cluster into eCat. Project Name: Automation Tool For Amazon Web services. Client: C oyote Logistics

This tool feature is to automate the data extraction from an SQL server and upload to Amazon S3, Launch EMR Cluster with provided EC2 Instances type, Setup tunneling between client machine and ec2 instances and direct upload the PIG/HIVE script to the launch EC2 Instance, Port forwarding so that Job tracker (Progress of the EMR) can be seen on user machine. Download the logs and output of EMR back to Amazon S3 and local/client machine.

● Written this automation tool by using the aws sdk of java.

● Created Job Flows to run elastic mapreduce.

PERSONAL INFORMATION:

Name:G.HARI KIRAN

Gender:MALE

DOB:04 05 1992

Email id:h ************@*****.***

Contact:077********.

Contact this candidate