Data Science Portfolio

---

Repository containing portfolio of data science projects completed by me for academic, self learning, and hobby purposes. Presented in the form of iPython Notebooks.

Contents

Natural Language Processing

• Spam detection: A classification model to detect whether a text message is spam or not (aka Ham) using simple logistic regression in PyTorch.

• Sentiment analysis: A classification model to predict the sentiment of text (Positive or Negative) using RNN, LSTM deep learning models in PyTorch. Used dropout to get rid of Overfitting.

• Autocorrection: A Probabilistic language model with NLP to implement an auto-correct system with the help of minimum edit distance algorithm. (for example, if you type the word 'dys', chances are very high that you meant to write 'days').

• Autocompletion: A Probabilistic language model which assigns the probability to a sequence of words, in a way that more likely sequences receive higher scores. (for example, if you type 'hey how are', then you can find a word 'you' that has the highest probability to match with the sentence 'hey how are'.).

• Identificarion of duplicate Questions: A model to detect duplicate questions in the websites like Quora, Stack Overflow, using deep learning techniques, including Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM).

• Neural Machine Translation: A sequence-to-sequence (seq2seq) model to translate the given sentence from German to English. Both the input and output are sentences. In other words, these sentences are a sequence of words going in and out of a model. This is the basic idea of Sequence-to-Sequence modeling which contains 2 different recurrent neural network (RNN) models knowns as an encoder and a decoder.

• Text Summarization: An attention model to summarize the text based on deep learning techniques.

Deep Learning

• Covid-19 Image Classification: An Image classsification model to classify the input image in three categories (If lungs are Normal, infected due to covid, or viral Pneumonia) using pretrained model of ResNet34.

• Christmas Multiclass Classification: A multiclass Image classification model to classify the images in Nine categories (Christmas Cookies, Christmas Presents, Christmas Tree, Fireworks, Penguin, Reindeer, Santa, Snowman) using AlexNet pretrained model.

Machine Learning

Regression
Linear Regression Support Vector Regression (SVR) Decision Tree Regression Random Forest Regression CO2 emission prediction CO2 emission prediction CO2 emission prediction CO2 emission prediction Ecommerce Customers Ecommerce Customers Ecommerce Customers Ecommerce Customers House Price Prediction House Price Prediction House Price Prediction House Price Prediction

Classification

Logistic Regression Naive Bayes Support Vector Machine (SVM) Kernel Support Vector Machine (K-SVM) Decision Tree Classification Random Forest Classification K-Nearest Neighbors (K-NN) click_on_Ad_or_not click_on_Ad_or_not click_on_Ad_or_not click_on_Ad_or_not click_on_Ad_or_not click_on_Ad_or_not click_on_Ad_or_not Kyphosis project Kyphosis project Kyphosis project Kyphosis project Kyphosis project Kyphosis project Kyphosis project