This blog post aims to provide one of the approaches used to solve the Kaggle problem of detecting the Partial Discharge (PD) pattern in the medium voltage power line signals.

The blog is structured in the following way:

- Business Problem
- Constraints
- Machine Learning problem statement
- Data overview
- Performance metric
- Exploratory Data Analysis
- Data pre-processing
- Featurization
- Final model
- Results
- Conclusion
- Future Works
- Profile

Overhead power line signals run for 100s of kilometers transferring power from one region to another. These distances make it difficult and expensive to manually inspect for any damages caused to the power lines. The damages could be…

This article aims to provide a very brief introduction to the basic intuition behind Dropouts in Neural Network.

When the Neural Network (NN) is fully connected, all the neurons in the NN are put to work, irrespective of them being useful in the particular task.

This article aims to provide a brief introduction as to how Matrix Factorization techniques can be used to create word vectors for a given document. This is achieved using Truncated Singular Value Decomposition (SVD) technique.

The basic formulation of SVD is as shown in Figure 1.

This article attempts to provide a brief introduction to the co-occurrence matrix and its implementation in python.

Given a document with a set of sentences in it, the co-occurrence matrix is a matrix form of representation of this document. To core idea of the co-occurrence matrix is to check if a particular word appears in the context of a focus word.

Let us take an example to understand this better. Let us consider a document containing two sentences S1 and S2 as shown in Figure 1.

This article is the continuation of Matrix Factorization for Collaborative Filtering.

Here, we take an example of user-item matrix A and try to understand how the factorization and prediction take place. The implementation of this is done in python.

Matrix A contains all users represented in rows and all movies represented in columns as shown in Figure 1.

In this blog post, we try to understand the basic intuition behind the use of Matrix Factorization for Collaborative Filtering in the Recommendation Systems.

The core idea behind Collaborative Filtering is that the users who have agreed in the past tend to agree in the future.

Let us understand this by an example. Let users be represented by vector U, and movies are represented by vector I as shown in Figure 1. …

This blog post tries to give a brief introduction as to how Matrix Factorization is used in K-means clustering to cluster similar data points.

The primary objective function in K-means clustering is given by:

Engineer, ML Enthusiast