Niloy Kumar Kundu

Niloy Kumar Kundu

Junior Machine Learning Engineer at Nodes Digital Limited

Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Posts

Understanding Matrix Multiplication with NumPy

4 minute read

Published:

Matrix multiplication can be quite confusing, especially when using the versatile np.dot() function in NumPy. In this blog, we’ll dive into the three main types of matrix multiplication: vector-vector, vector-matrix, and matrix-matrix operations. We’ll clarify how these operations work and provide examples to enhance your understanding.

10. Understanding Word Embeddings

3 minute read

Published:

In Natural Language Processing (NLP), a word embedding is a representation of a word in a continuous vector space. This representation is typically a real-valued vector that encodes the meaning of the word, such that words closer together in the vector space are expected to have similar meanings.

9. Understanding LSTM Networks

5 minute read

Published:

In Recurrent Neural Networks (RNNs), one of the major challenges is the vanishing gradient problem. To address this, we use more advanced architectures like Long Short-Term Memory (LSTM) networks and Gated Recurrent Units (GRUs). Today, we’ll focus on LSTM networks.

8. Problems in Simple RNNs

3 minute read

Published:

In a Recurrent Neural Network (RNN), the process starts with forward propagation, followed by backward propagation (Backpropagation Through Time, or BPTT). During backward propagation, the network’s weights are updated over time to minimize the loss function. However, simple RNNs encounter significant challenges known as the vanishing gradient problem and the exploding gradient problem.

5. Word2Vec

3 minute read

Published:

1. Lack of Semantic Information

Semantic information refers to the meaning and relationship between words in a sentence or document. Traditional methods like Bag of Words (BoW) and TF-IDF focus solely on the frequency of words and ignore the context in which they appear. This means they don’t capture the meaning of the words or how they relate to each other.

4. Understanding TF-IDF

2 minute read

Published:

TF (Term Frequency)

Term Frequency (TF) is a measure of how frequently a word appears in a sentence, normalized by the total number of words in that sentence. It is calculated as:

3. Bag of Words in NLP

3 minute read

Published:

The Bag of Words (BoW) model is a fundamental technique in Natural Language Processing (NLP) used to extract features from text data. It helps in representing text in a numerical form, which is essential for many machine learning algorithms. In this post, we’ll explore how the Bag of Words model works, how to implement it, and some of its limitations.

2. Stemming and Lemmatization in NLP

2 minute read

Published:

In Natural Language Processing (NLP), reducing words to their root form is an essential step for various tasks like text analysis and classification. Two common techniques for this are Stemming and Lemmatization. Though they serve a similar purpose, they differ in their approach and results. In this post, we’ll explore both techniques and discuss when to use each one.

1. Tokenization in NLP

2 minute read

Published:

When dealing with text data, tokenization is a crucial step. It involves breaking down a text into smaller components, such as words or sentences, to prepare it for further analysis. In this post, we’ll explore how to handle tokenization using the Natural Language Toolkit (NLTK), an open-source library that simplifies various NLP tasks.

A Comprehensive Roadmap to Mastering Natural Language Processing

3 minute read

Published:

Natural Language Processing (NLP) is a rapidly evolving field with a broad spectrum of techniques and technologies. Whether you’re a beginner or looking to deepen your knowledge, this roadmap will guide you through essential stages of NLP. Here’s a structured path to mastering NLP:

portfolio

projects

Agri Advisory

A multilingual AI-driven platform that allows users to ask agriculture-related questions in text or audio (Bangla/English). It retrieves relevant information from a vector database and generates detailed responses using a Large Language Model (LLM), providing answers in both text and speech formats.

ML Backend Framework Development

Developed a scalable machine learning backend framework for efficient data processing, model deployment, and API services, enabling seamless integration of ML models into production environments.

NDL Research BOT

An AI-powered tool that processes and analyzes uploaded research papers in PDF format. Using a custom-built Large Language Model (LLM), the bot extracts key insights, providing quick summaries and answers to specific questions, saving time and enhancing research efficiency in our agriculture startup.

Water Height Estimation in Rice Fields

Developed a computer vision-based system combined with regression models to accurately estimate water height in rice fields. This solution automates the measurement process, improving irrigation management by providing real-time, precise water level data for better resource allocation and crop management.

publications

Paper Title Number 4

Published in GitHub Journal of Bugs, 2024

This paper is about fixing template issue #693.

Recommended citation: Your Name, You. (2024). "Paper Title Number 3." GitHub Journal of Bugs. 1(3).
Download Paper

talks

teaching

Teaching experience 1

Undergraduate course, University 1, Department, 2014

This is a description of a teaching experience. You can use markdown like any other post.

Teaching experience 2

Workshop, University 1, Department, 2015

This is a description of a teaching experience. You can use markdown like any other post.