kalgishah02 / snaploc

SnapLoc is a product that does automatic image classification and spatio-temporal analysis in order to recommend the places of interest in a new city. The packages that I have used for creating the product are Python(Pandas, NumPy, Shapely, Keras, Leaflet) and TensorFlow

SnapLoc

SnapLoc — Personalized recommender for Points of Interest in a city

SNAPLOC is a product that I have created as a final project of a data science bootcamp, Metis. It does automatic picture classification and spatio temporal analysis in order to recommend the places of interest for traveling in a new city. I have been a traveler to a new city and felt a need to know about the most interesting places to explore near me and have looked at photos of other people and said I want to go there to capture this beautiful picture of bear in dawn, if feasible.

Folder named Code contains three files:

1. Flickr_getimages - Code to download geo-tagged images and it's metadata from Flickr
2. Image_Classification - Code to do the VGG classification of images
3. Spatial_clustering - Code for doing spatial clustering using DBSCAN and the cluster visualization & ranking.

Folder named Data contains metadata of images downloaded for San Francisco from Flickr as well as tables created for analysis of clusters of images formed.