Data Science Portfolio

Here are some of my best Data Science Projects. I have explored various machine-learning algorithms for different datasets. Feel free to contact me to learn more about my experience working with these projects.

Portuguese Bank Marketing Analysis

Skills used: Python, Pandas, SKlearn, Matplotlib

Project Objective: In this project I worked with real world data which is related with direct marketing campaigns (phone calls) of a Portuguese banking institution. The classification goal is to predict if the client will subscribe a term deposit.

Quantifiable result: We could classify whether client has subscribed the term deposit or not 86% accuracy.

Used Logistic Regression to classify whether term deposit is subscribed or not.
Classified subscriotion depending on selected important features.
Explored a machine learning approach to marketing analysis.

Orthopedic Patients Prediction

Skills used: Python, Pandas, SKlearn, Matplotlib

Project Objective: In this project we are provided with multiple instances of orthopedic parameters according to that I classified whether the patients are Normal or Abnormal.

Quantifiable result: We could Classify the type of tumor resulting in 89% accuracy using NB algorithm)

Used K-Nearest-Neighbors and GaussianNB to classify Normal or Abnormal paitents.
Analysed different parameters to classify patients.
Explored a machine learning approach to Medical Analysis.

Employee Compensation

Skills used: Python, Pandas, SKlearn, Matplotlib, Clustering, Elbow

Project Objective: The San Francisco Controller’s Office maintains a database of the salary and benefits, paid to City employees since fiscal year 2013. This data is summarized and presented on the Employee Compensation.

Quantifiable result: We have grouped employees with 3 Clusters.

Used KMeans Algorithm
Used Elbow method to get the optimal number of cluster.
Explored an Unsupervised learning approach to create data group.

Talking Data Fraud Detection

Skills used: Python, Pandas, SKlearn, Matplotlib, XGBoost, Ensemble technique

Project Objective: In this problem, we will use the features associated with clicks, such as IP address, operating system, device type, time of click etc. to predict the probability of a click being fraud.

Quantifiable result: We could cpredict whether a given click resulted in a download or not 88% Accuracy using XGBoost.

Used the XGboost Classifier algorithm to classify the Pepole who downloaded the app after watching the advertisement and identify the click fraud
Compared predictive performance by fitting a BaggingClassifier model to the data
Selected best model based on train and test performance
Explored Ensemble technique.

Hand Writing Digits Recognition

Skills used: Python, Keras, Tensorflow

Project Objective: Mnist is a dataset with 70,000 of handwritten digits. Here I have predicted the mnist digits by using Convolution Neural Network.

Quantifiable result: We could train the Convolutional Neural Network to attain a accuracy of 98% using 5 epochs.

Added multiple convolution and pooling layers
Training model on basis of Mnist data
Data Source: MNIST