Home

Achyuthuni Sri Harsha

Who am I

I am currently working as a Senior Business analyst at Tesco Bengaluru. In the space-range and merchandising domain, I am trying to solve business problems using statistical concepts and data science. I am also pursuing MSc in Business Analytics (part-time) from Imperial College London.

I am an alumnus of the Indian Institute of Management, Bangalore in Business Analytics and Intelligence. BAI is an executive on-campus one-year program with a focus on statistics, business and technology. 'Building R&R contests for one of India's largest life insurance company' which is our team's capstone project got the 'Highly commended project' award.

Previously I worked as a decision scientist at Mu-Sigma and have pursued my bachelors at Amrita Vishwa Vidyapeetham in Mechanical engineering. I have experience working with industry giants and Fortune 500 companies in retail, pharma, insurance and IT industries.

Why this blog

Sometimes, unless I do something, I don't completely understand it. After learning something new, especially in the space of data science, I try to solve a simple problem using what I just learned. This blog is a cumulation of all my notes from such learnings.

What else

I try to use problems and issues from my everyday life and try to solve them using statistical/data science lens. I have been attempting to publish some of these everyday problems which can be solved using data science as papers and conference articles, some of which are:

  1. Personal analytics: Time management using Google maps (presented in ICSADADS-2020)
  2. Personal analytics: Selection of a phone using AHP

Other publications include:

Interested to learn business analytics with me?

Checkout the blog

Business Analytics

Data Processing

Web scraping

  1. Handling Google Maps Location Data (in-time problem)
  2. Class size paradox and web scraping (using Amrita University placement data)

Data cleaning and imputation

  1. Null value imputation using KNN (mtcars data)

Exploratory Data Analytics

  1. Univariate Analysis (in-time problem)
  2. Multivariate Analysis (in-time problem)
  3. Multicollinearity (in-time problem)
  4. Time Series EDA (in-time problem)

Factor analysis

  1. Curse of dimensionality
  2. Exploratory factor analysis

Inferential data analytics (Hypothesis testing)

  1. z-test and t-test (in-time problem)
  2. ANOVA test (smart cities data)
  3. Chi-Square Goodness of fit test (in-time problem)
  4. Chi-Square test of independence (smart cities data)

Prediction algorithms (Supervised learning)

  1. Classification
    1. Logistic Regression (using Titanic data set)
    2. CHAID decision trees (using Titanic data set)
    3. CART classification (using Titanic data set)
  2. Regression
    1. Part and partial correlation
    2. Linear regression (Boston housing problem)
  3. Machine Learning
    1. Handling Imbalanced Classes

Prescriptive Analytics (Optimization)

  1. Linear Programming and Sensitivity analysis (basic)
  2. Inventory planning model (with CPLEX code)
  3. Gradient descent for non-linear optimization (Adoption of a new product)
  4. Analytic Hierarchy Process for multi-criterion optimization (Selecting a phone)

Reinforcement Learning (Stochastic modelling)

  1. Recommendation system (associate mining)
  2. Markov Chains introduction (Customer Lifetime Value)

Time series forecasting

  1. Introduction to stationarity
  2. Stationarity hypothesis tests (in-time problem)
  3. Forecasting using ARIMA (in-time problem)

Clustering

  1. Hierarchical Clustering (Market segmentation using wine data)
  2. K means clustering (Customer segmentation using credit card data)

Deep Learning

  1. Artificial Neural Network – part 1
  2. The math behind ANN

Projects

  1. Personal analytics: TIme management using google maps
  2. Supply chain analytics: Purchase order forecasting
  3. Optimization of Rewards and Recognition's for a reputed insurance company using predictive and prescriptive analytics

Other interesting posts

  1. Review on IIMB Business Analytics and Intelligence course
  2. Why are basics important in data sciences
  3. Boy-girl paradox

Python tutorials ( Authored by Tamoghna Saha )

  1. Getting started with python
  2. A Not-so-Quick-but-Conceptual guide to Python  | Intermediate | Part 1
  3. A not so quick but conceptual guide to python notebook intermediate | part-2
  4. All you ‘really’ need to know | Python Notebook | Advanced – Pandas

Papers and publications

  1. Parametric Study of Cantilever Plates Exposed to Supersonic and Hypersonic Flows (published in IOP Material Science)
  2. Purchase order forecasting (under publication- Accepted for publication in NCMLAI 2020)