Follow Us On

Youtube Google Plus linkedintwitterfacebook

Analytics And Big Data

heading

last update : 12/01/2016

Certified Apache Mahout

Event Date Country City Days Price  
    No upcoming event date found      
noc

 

COURSE OBJECTIVE

Scalable, commercial-friendly machine learning for building intelligent applications

This course will introduce you to the basic blocks of machine learning, and where Mahout fits in. We will majorly be looking at recommendation systems, what are their types, how to choose a similarity algorithm, and a typical design of a recommendation system. We will be exploring many examples of recommendations in the real world. We will also be running recommendations for a pretty large dataset on Hadoop over Amazon Elastic MapReduce.We will also be looking at the basics of Clustering, a typical Clustering algorithm, with an example.Lastly, we will look into the basics of Classification, it's types and some examples.

The need for machine-learning techniques like clustering, collaborative filtering, and categorization has never been greater, be it for finding commonalities among large groups of people or automatically tagging large volumes of Web content.

Hardware and Softwares Required for this training:

Delegates attending this course are required to carry a laptop with below configuration. Usually we suggest a 64 bit machine and not a 32 bit machine. However you can still try with a 32 bit machine but we do not gaurantee if the programs will run without any error. 

Windows userWindows 7 – 64 bit OS, Min 4 GB RAMVMWare Player 5.0.0Linux  VM– Ubuntu 12.04 LTS : {Unicom will be providing the VM(virtual machine)}Eclipse 3.6+Putty – For opening Telnet sessions to the Linux VMWinSCP – For transferring files between Windows and Linux VM

Linux/Mac Users (preferably a 64 bit machine): Min 4 GB RAM Eclipse 3.6+JDK 1.6 or higher installed on your machineSSH installed

Systems with said configuration can be arranged through us with an additional cost.

 

About

The Apache Mahout project aims to make building intelligent applications easier and faster. Mahout co-founder Grant Ingersoll introduces the basic concepts of machine learning and then demonstrates how to use Mahout to cluster documents, make recommendations, and organize content.

Apache Mahout is an Apache project to produce free implementations of distributed or otherwise scalable machine learning algorithms focused primarily in the areas of collaborative filtering, clustering and classification, often leveraging, but not limited to, the Hadoop platform.

Three specific machine-learning tasks that Mahout currently implements.    Collaborative filtering    Clustering    Categorization

The need for machine-learning techniques like clustering, collaborative filtering, and categorization has never been greater, be it for finding commonalities among large groups of people or automatically tagging large volumes of Web content.

 

Agenda
  • Intro to Machine Learning
  • Intro to Apache Mahout
  • Recommendation Engine
  • Clustering
  • Classification
  • Intro to recommendation systems
  • Content Based
    • Collaborative filtering
    • User based
    • Nearest N Users
    • Threshold
    • Item based
  • Mahout Optimizations
  • An overview of a recommendation platform
    • Similarity measures    
    • Manhattan distance
    • Euclidean distance
    • Cosine Similarity
    • Pearson's Correlation Similarity
    • Loglikihood Similarity
    • Tanimoto
  • Evaluating Recommendation engines
    • Online
    • Offline
  • Intro to Clustering
    • Common Clustering Algorithms
    • K-means
    • Fuzzy K-means, Mean Shift etc
    • Representing data
    • Feature Selection
    • Vectorization
    • Representing Vectors
  • Intro to Classification
    • Examples
    • Basics
    • Common Algorithms
  • Mahout on Hadoop
  • Apache Mahout & Myrrix

 

Instructor Bio

The Trainer has around 10 years of tech industry experience across global services companies such as iGate & Sapient, huge web based product development companies(AOL) & a couple of online product startups. He has been instrumental in setting up and leading the data platform team at Qyuki Digital Media, a Shekhar Kapur & AR Rahman social media creativity initiative. Apart from vast experience in Java, He has around 2 years of big data mining & analytics experience involving technologies such as Hadoop, Apache Mahout, Cassandra & other machine learning packages such as Weka. He is passionate about the Hadoop ecosystem, personalization & recommendation systems.He has a passion for explaining seemingly difficult concepts to people in a simple way and has conducted multiple corporate/online trainings for Java, Hadoop/MapReduce, building recommendation systems & Cassandra.

Submit your details to download the brochure:

First Name *:

Last Name *:

Email *:

Phone Number *:

Job Title:

Organisation:

Comments:

  Type the characters you see in the picture below *:

 

 

FAQ

What is Course Pre-Requisites?The participants should have basic understanding or knowledge of java and linux

How can I make payment?Payment can be made via Cheque / DD / Online Funds transfer / Cash Payment.Cheque should be drawn in favour of "Unicom training and Seminars Pvt Ltd" payable at BangaloreNEFT Payment:Account Name: UNICOM Training & Seminars Pvt LtdBank Name : State Bank of IndiaBank Address: Ground Floor, K V Plaza, Green Glen Layout, Outer Ring Road, Bangalore.A/c Number : 31729010535IFSC : SBIN0012706A/c Type: CurrentWhat is Course timing?0900 – 1700 each dayWhat is the course Fee?2 Days Course - Rs 18,000 + 12.36% (Service Tax)Whom do I contact for more details?+91-9538878795 or contact@unicomlearning.com

QUICK QUERY

 
navigation div
navigation div

Contact Us(India)

Shanmugha Arcade,

3rd Floor, 39,

NGEF Lane,

Indira Nagar 1st Stage,

Bengaluru - 560038,

Karnataka, India.

Telephone: +91-9538878795, +91-9538878799, +91-8025257962

E: contact@unicomlearning.com

Contact Us(UK)

OptiRisk R&D House

One Oxford Road

Uxbridge

Middlesex

UB9 4DA

UNITED KINGDOM

E: contact@unicom.co.uk

© 2018 All Rights Reserved