Somjit Bhowmick's repositories
Predictive-Maintenance-Industrial-IOT
Illustrating a typical Predictive Maintenance use case in an Industrial IoT Scenario. By using Statistical Modelling and Data Visualization we attempt to performance Failure Analysis and Prediction of crucial industrial equipments like Boilers, Pumps, Motors etc. so that necessary actions can be taken by the management for their repair, servicing and optimal performance.
Facebook-Friend-Recommendation
This is a friend recommendation systems which are used on social media platforms (e.g. Facebook, Instagram, Twitter) to suggest friends/new connections based on common interests, workplace, common friends etc. using Graph Mining techniques. Here, we are given a social graph, i.e. a graph structure where nodes are individuals on social media platforms and a directed edges (or 'links') indicates that one person 'follows' the other, or are 'friends' on social media. Now, the task is to predict newer edges to be offered as 'friend suggestions'.
Min-Cost-Network-Flow-Lemon
A Simple C++ Implementation of the Lemon Optimization Library to Solve a Minimum Cost Flow problem in a given Graph Network with Supply/Demand Values of Nodes and Capacity, Unit Cost of Flow for each Edge.
BERT-Question-Answering
A study on encoding english sentences to tensorflow vectors or tensors using pre-trained BERT model from the Hugging Face Library.
Human-Activity-Recognition
This project is to build a model that predicts the human activities such as Walking, Walking Upstairs, Walking Downstairs, Sitting, Standing or Laying using readings from the sensors on a smartphone carried by the user.
IMDb-Sentiment-LSTM
A study of the use of Long Short Term Memory (LSTM) for the sentiment classification of movie reviews on the well-known IMDb website.
MNIST-Classification-Keras
A simple study on the use of Keras framework (with Tensorflow background) for a simple handwritten number image classification task with Deep Neural Networks.
MNIST-Classification-Tensorflow
A simple study on how to use Tensorflow platform (without Keras) for a simple number classification task using a Neural Network.
Netflix-Movie-Recommendation
A case study of the Netflix Prize solution where, given anonymous data of users and the ratings given to movies, the objective to provide recommendations to users for movies which they would like, based on their past activity and taste.
NLP-CaseStudy-Quora-Question-Similarity
An application of NLP and classical ML algorithms to an interesting real-world use case of predicting similarity between two questions on Quora. This allows the platform to combine similar questions into one and combine their answers to avoid duplication and unnecessary confusion.
NLP-Star-Trek-Scripts
Using digital form of the actual scripts of the 'Star Trek' science fiction series to perform interesting NLP tasks and answering some questions on Topic Modelling, Character properties and the plot as a whole.
NYC-Taxi-Demand-Prediction
This is a Time Series Forecasting and Regression solution to project the no. of pick-ups at and around a given region at a given time in the city of New York, USA.
Basic-Website-Design-Flask
A very basic website designed using custom HTML templates and CSS stylesheets and rendered as a microservice in Python using the popular Flask library.
COVID-19-Optimal-Resource-Allocation_and_Request-Classification
A solution developed to Map essential COVID-19 Relief resources to the needy across a city in the most cost-optimal way, and also to classify incoming SOS messages from those in need of help, for organizational and lesser response times.
Music-Generation-LSTM
A simple application of an LSTM Network to generate original Folk-style music which are aesthetically pleasing for the listener.
NLP-StackerOverflow-Tag-Prediction
A multi-class classification problem where the objective is to read a question posted on the popular reference website, StackOverflow and predict the primary topics it deals with, i.e. tags which the question will be associated with.
Data_Science-EDA
A collection of useful implementations to perform EDA on a new dataset in order to understand preliminary patterns in the dataset and gain a high-level grasp of the dataset using plots and visualizations.
DCGAN-Gradient-Tape
A study of the use of the Tensorflow GradientTape class for differentiation and custom gradient generation along with its use to implement a Deep-Convolutional Generative Adversarial Network (GAN) to generate images of hand-written digits.
Dimensionality-Reduction-PCA-t-SNE
Implementation of Principal Component Analysis (PCA) and t-Distributed Stochastic Neighbour Embedding (t-SNE) as Dimensionality Reduction techniques using MNIST Dataset.
DS-kNN-K-Fold-CV
An implementation of the **k-Nearest Neighbours Classification Algorithm** using some popular demo datasets.
DS-Linear-Regression
A simple Linear Regression model using the Boston Housing Prices Dataset
DS-Logistic-Regression
A simple implementation of the Logistic Regression Classifier on the Breast Cancer Dataset with L1 regularization and GridSearch for hyperparameter tuning.
Microsoft-Malware-Detection
A multi-class classification problem where the task is to classify a file to one of 9 types of Malware usually found in a Windows system, using information from the raw data and metadata of the file.
MNIST-Classification-CNN
A simple study on the use of CNNs for a simple handwritten number image classification task using the Keras framework (with Tensorflow background).
NLP-CaseStudy-Amazon-Fine-Foods-Review
Efficient Sentencing Encoding and Vectorization techniques with customer reviews on a product page of the popular E-Commerce website, Amazon using proven NLP techniques for the purpose of sentiment analysis.
NLP-CaseStudy-Personalized-Cancer-Diagnosis
In this problem statement, a sequence of genetic mutations and clinical evidences, i.e. descriptive texts as recorded by domain experts are used to classify the mutations to conclusive categories, to be used for diagnosis of the patient.
Search-Engine-Stackoverflow-QnA
An application to store a collection of questions and answers data found on Stackoverflow in an index using Elastic and perform a text search on the stored Q&A's based on semantic meaning.
Self-Driving-Car
An implementation of a simple self-driving car control using the image feed from a single dashcam.
Time-Series-LSTM
A detailed study on the use of Long Short Term Memory (LSTM) units for time series forecasting with climate data.