kiran karkera
Data scientist
Dex.sg
location_on Singapore
Member since 8 years
Share
kiran karkera
Specialises In
In my data science career, I have had the opportunity to wear different hats.
As a consultant, I help customers in their journey to use data for decision making. I work with them to understand the strengths and weaknesses of the data they have and advise them on the possible choices for analytical solutions. I collaborate with them to implement machine learning solutions and productize the solution to aid their experts in decision making.
As a product lead, I have led teams in building products that use machine learning and NLP, in the domain of customer experience and product sales recommendations.
As a machine learning engineer, I have built ML solutions and contributed to open source ML tool-kits in the Clojure ecosystem such as Cortex and Clj-ml. I’ve participated in a few ML competitions, and I have been placed among the top 1% of data scientists at Kaggle.com (as of Dec 2013).
I have authored a book ‘Building Probabilistic Graphical Models in Python’, which is a gentle introduction to Graphical Models that have applications in Machine Learning and NLP.
I have filed two patents that help improve Femtocell Devices and Smallcell Device Management Server efficiency using machine learning approaches.
I’m a programmer with 19 years of product development experience, having donned the roles of a developer, team lead and architect in start-ups and product companies.
I have worked in varied domains such as Ad-tech, Customer Experience, Network and Device Management and FMCG. I have also worked with many cross functional as well as globally diverse teams.
-
keyboard_arrow_downIntroduction to Bayesian Networks
Shrutika PoyrekarData SientistEnvestnet | Yodlee
kiran karkeraData scientistDex.sgschedule 3 years ago
90 Mins
Workshop
Advanced
{ This is a handson workshop . The use case is Traffic analysis . }
Most machine learning models assume independent and identically distributed (i.i.d) data. Graphical models can capture almost arbitrarily rich dependency structures between variables. They encode conditional independence structure with graphs. Bayesian network, a type of graphical model describes a probability distribution among all variables by putting edges between the variable nodes, wherein edges represent the conditional probability factor in the factorized probability distribution. Thus Bayesian Networks provide a compact representation for dealing with uncertainty using an underlying graphical structure and the probability theory. These models have a variety of applications such as medical diagnosis, biomonitoring, image processing, turbo codes, information retrieval, document classification, gene regulatory networks, etc. amongst many others. These models are interpretable as they are able to capture the causal relationships between different features .They can work efficiently with small data and also deal with missing data which gives it more power than conventional machine learning and deep learning models.
In this session, we will discuss concepts of conditional independence, d- separation , Hammersley Clifford theorem , Bayes theorem, Expectation Maximization and Variable Elimination. There will be a code walk through of simple case study.
-
-
No more submissions exist.
