A model to predict total vaccination by given date using naive simple linear regression approach. This model is fitted on the raw dataset “country_vaccinations” which is gathered from the government official websites. The prediction is done with respect to one country “India”.

Implemented a Multiple Regression model in Machine Learning to fit the model based on features/ independent variables and dependent variable. The dataset used here is of “winequality-red” wherein 1599 rows x 12 columns.

Shown comparison of multiple and polynomial regression techniques by calculating each MSE value and determining which is better for wine quality determination. The dataset used here is of “winequality-red”.

A Decision Tree model to fit the model towards the regression to predict housing prices by using “melb_data.csv” dataset.

A Logistic Regression model in Machine Learning to fit the model to predict heart rate of a person based on age and BMI by using “Framingham.csv” dataset.

Implemented model for letter recognition problem that solved using SVM (Support Vector Machine) by using “letter-recognition.csv” dataset.

Analyze to Predicting Diamond prices using KNN (K Nearest Neighbours) Regression. By using different features from “diamonds.csv” dataset.

Implementation of K-Means Clustering to show clusters of different type of wines. Here dataset is used of “wines.csv”.

Neural Machine Translation (NMT) model to translate the sentences from English to Marathi language. This is basic level sequence to sequence (seq2seq) model using LSTM over 39961 pair of sentences using mar-eng-2/mar.txt dataset. The words predicted for most of the sentences are semantically correct.

Predicting answer to the questions in Hindi and Tamil by using RoBERTa model (Robustly optimized BERT pre-training approach) with the chaii - Hindi and Tamil Question Answering Dataset hosted by Google. For the model the Hugging face library has provided excellent documentation with the implementation.

Sentiment Analysis with Naive Bayes by analysing the sentiment of a given text or document and categorising the text/document into a specific class or category (like positive and negative). This is done by using NLTK movie review corpus labeled training data containing 2K movie reviews with sentiment polarity classification.

Simple Information Retrieval by firstly creating TF-IDF(Term Frequency-Inverse Document Frequency) Matrix and applying SVD on it and visualising those representation and using LSA(Latent Semantic Analysis) for Information Retrieval. Dataset used is Women’s Clothing E-Commerce.

Modeled a simple spam filter using naive bayes & labelled dataset of email obtained from Identify Spam using Emails dataset. Explained various preprocessing steps involved for text data followed by feature extraction & calssification model.

Build and trained a basic speech recognition model to recognise different words by using mini speech commands dataset From reading the audio files from the labels and converting waveforms to spectrograms with 84% model accuracy.

Text Summarisation technique that extracts text from a large amount of data such as from wikipedia articles obtained through URL by concepts of Web Scrapping by BeautifulSoup library and then processing and calculating the sentence scores to give the summary of the article.

Text classification starting from plain text files to train binary classifier to perform sentiment analysis on an IMDB dataset i.e Large Movie Review Dataset Implemented with an accuracy of 87.28 %.

Contrast Stretching (normalisation) - that attempts to improve the contrast in an image by 'stretching' the range of intensity values it contains to span a desired range of values, and Image Negative - an image where its lightest areas appear as darkest and the darkest areas appear as lightest.

Noise Removal using average filtering and median filtering used directly on pixels of an image. Mask is usually considered to be added in size so that it has a specific center pixel. This mask is moved on the image such that the center of the mask traverses all image pixels.

Morphological Operation Extract large image features - Dilation & Erosion - Dilation adds pixels to the boundaries of objects in an image, while erosion removes pixels on object boundaries.

Canny Edge Detection - an edge detection operator that uses a multi-stage algorithm to detect a wide range of edges in images.

Harris Corner Detection - Harris Corner Detector is a corner detection operator that is commonly used in computer vision algorithms to extract corners and infer features of an image.

Random sample consensus, or RANSAC, is an iterative method for estimating a mathematical model from a data set that contains outliers. The RANSAC algorithm works by identifying the outliers in a data set and estimating the desired model using data that does not contain outliers.

Hough Transforms - reading the image through Hough Transforms. The Hough transform is a feature extraction technique used in image analysis, computer vision, and digital image processing. The purpose of the technique is to find imperfect instances of objects within a certain class of shapes by a voting procedure.

Face Detection using Viola-Jones Algorithm which is both feature and template based, detection is very fast, simple to understand and implement, less data needed for training than other ML models, no rescaling of images needed (like with CNNs), much more interpretable than contemporary models.

Embedded assistive stick for visually impaired people, made up of HC-SR04 ultrasonic sensor and buzzer for its sound signal. Using this smart blind stick, a visually impaired person can walk without anyone's help. The smart blind stick automatically detects the obstacle in front of the person and gives him a response to the person by warning sound.

To maintain the ‘Blood Cold Chain’ one needs to regulate the temperature and provide alerts when it breaks constraints so that precaution is taken much earlier. Sudden change in temperature is also detected and alerted. Prediction of future 20 min temperatures makes this system more easy to regulate and prevent future anomalies.

COVID- 19 patients are being held in isolation in various hospital wards or quarantine facilities, making it difficult and dangerous for medical staff to manage their live data. The goal of this project is to make it easier to monitor a patient's health while they are in quarantine from a single web server.

Arduino-based system for detecting landslides. This project has the ability to track the earth's vibration and outputs a light signal and buzzer sound. In hilly or mountainous places, this device might be installed close to landslide-prone zones.

The Temperature Monitoring robot, as the name implies, is an autonomous vehicle that avoids the obstacle coming at its surface and simultaneously detect the temperature and shares that data to the user. This robot can be used for the temperature monitoring at dangerous environments, where human penetration could be fatal.

The Voice Command robot, as the name implies, is an autonomous vehicle. The objective is simple-if the sensors encounter an obstacle they move back and turn left or right as the case may be. Voice Command is given through google assistant. Obstacle avoiding robots can be used in almost all mobile robot navigation systems. They can be used for household work like automatic vacuum cleaning.

The line follower robot car, is an autonomous vehicle that follows a visible line imprinted in the surface. This visual line represents the path that the line follower robot will take. Similar like these giant line follower robots are employed in industries to assist in the automated production process. They're also employed in military applications, humanitarian aid, and delivery services, among other things.

Making a robot that will follow people or items is the goal of this project. To manoeuvre the robot in this, infrared sensors are mounted. The Arduino Uno microcontroller serves as the robot's brain. DC gear motors are used to power the robot and steer the wheels. This machine can be used for a variety of tasks, from small-scale tasks like moving goods around the house to large-scale tasks like moving big objects during construction or in factories.