Summary
Summary
Embedded Files
As I developed my machine learning career by doing several courses online, I understood that data is big aspect of Artificial Intelligence. Preprocessing, Visualising and Analysing data by performing exploratory data analysis is just a start to what can be done with data science. Capstone project, accredited by IBM in IBM Data Science Professional Certificate allowed me to display my skills in data science and creating models that are applicable in real life.
As I developed my machine learning career by doing several courses online, I understood that data is big aspect of Artificial Intelligence. Preprocessing, Visualising and Analysing data by performing exploratory data analysis is just a start to what can be done with data science. Capstone project, accredited by IBM in IBM Data Science Professional Certificate allowed me to display my skills in data science and creating models that are applicable in real life.
Capstone Project- Venue Based Similarity Measures
Capstone Project- Venue Based Similarity Measures
Language:
Language:
Python
Tools & Libraries
Tools & Libraries
Pandas
Numpy
Matplotlib
Seaborn
Folium
FourSquare API
Geolocator API
Summary:
Summary:
Compared two Neighbourhoods-East Central London & Birmingham based on Venues Categories
Performed Data Cleaning, Visualisation & Machine learning on the obtained Data
Data Cleaning by using Pandas tool
Data Visualisation using Matplotlib
Machine Learning using Scikit-Learn
Used unsupervised learning algorithm of K-means clustering to form inside City Clusters - ECL: 6, Birmingham: 2
Discussed the application fo clustering with venues categories for each neighborhood
The Pitch
The Pitch
Venue-Based Similarity Measures Pitch.pptxVenue-Based Similarity Measures in East-Central London & Birmingham Pitch
House Sales Analysis and Prediction in King County, USA
House Sales Analysis and Prediction in King County, USA
Language:
Language:
Python
Tools & Libraries
Tools & Libraries
Pandas
Numpy
Beautiful Soup
Scikit Learn
Summary:
Summary:
Used extensive python library to perform Exploratory Data Analysis & Visualisation
Developed a Multiple-Linear regression model with accuracy of 75% to correctly predict house prices in King County, USA.
Used train,cross validation, test-split and changing model from Linear to Polynomial, increasing the accuracy of the model to 94%
Best Binary Classification Model- Classifying customers according to their Loan status
Best Binary Classification Model- Classifying customers according to their Loan status
Github:
Github:
Language:
Language:
Python
Tools & Libraries
Tools & Libraries
Pandas
Numpy
Scikit-Learn
Summary
Summary
Performed Data-visualisation, preprocessing and normalization to ensure data is suitable for analysis
Developed different classifer model tfr comparisonwith K-nearest nighbour, Decision tree, Support Vector Machine (SVM) , and Logistic regression algorithms.
Compared models with F1 score, Jaccard, and log loss methods and concluded the best model.
Page updated
Google Sites
Report abuse