![profile](images/image1.jpeg)
Nihar Jamdar
Data Analyst at Merkle
About Me
Nihar Jamdar
Data Analyst at Merkle
With over 3 years of experience, I specialize in leveraging advanced analytics and machine learning to drive actionable insights. Proficient in Python, SQL Server, and TensorFlow, I have a proven track record in developing predictive models for diverse industries. Skilled in cross-functional collaboration and data-driven decision-making, I excel in delivering impactful solutions to complex challenges.
Birthday: 28 Nov 1997
Age: 25
Email: nbjamdar@gmail.com
Phone: +91 9769069453
City: Mumbai, India
Languages: Marathi, Hindi, English
Machine Learning
Python
Natural Language Processing
Deep Learning
Databricks
SQL Server
Work Experience
8/2021 - Present
Merkle
Efficiently produced and delivered performance reports at various intervals (weekly, biweekly, monthly, quarterly) using SQL Server, Tableau & Excel Pivot, cutting processing time by 50%
Developed multiple ML models for regression and classification, consistently achieving 80-90% accuracy by aligning them with problem statements, ultimately leading to increased customer acquisition and retention
Delivered technical expertise to numerous clients during aligned discussions to address their specific business inquiries
Latest Blogs
![blog](images/portfolio/tesnorflow_.jpg)
Multi-class Text Classification using TensorFlow
In this blog, I have created multiclass text classifier by computing their weights and then assigning higher weights to label with lesser data points after that using tenorFlow with weighted cross-entropy func to execute results.
![blog](images/portfolio/nlp.jpg)
TF-IDF (Term Frequency and Inverse Document Frequency)
Tf-Idf- is one the Natural Language processing techniques which is used to convert words into vectors, so here i have tried to explain it will basic example with mathematical formula.
![blog](images/portfolio/netflix-logo.png)
Netflix Data Analysis using Python
Performed Exploratory Data Analysis on Netflix Data with help of basic visualization techniques using Python libraries like Seaborn,Matplotlib etc.
![blog](images/portfolio/mnist.jpeg)
2-D Visualization using Principal Component Analysis (PCA) on MNIST dataset
Principal Component Analysis is one of the technique which is used for reduction of dimension where it converts d-dimension into d' dimension where d'< d. So here i have created a blog by applying PCA on MNIST dataset to convert 784-dimensions to 2-dimension.
![blog](images/portfolio/multi.jpg)
What is MultiCollinearity and how to resolve it
If any of our Independent Feature (xi, xj) is internally co-related more than 90%. So how we can solve multicollinearity issues practically using Python and statistics.
![](images/portfolio/pesantren-alandalus.com.png)