CUBE

BRIEF

In our world there is lots of changes happening to our earth, either it is manmade or by nature, have you ever gazed out the window of a flying airplane and wondered what happened here? what is there? And what would happen through years in here?, So we have built a responsive web application, uses real time precise location of the device, send it to the server, and send back the data associated with the location to entertain you with NASA’s way of entertainment (which is information about that place, predict the future for that place ) in addition with using data visualization and machine learning.

WHAT WILL BE DONE?

• RESPONSIVE WEB APPLICATION, TO WORK ON ALL PLATFORMS, AND MAKE THE APPLICATION EASY TO USE.
• GOOGLE MAPS API, TO RETURN THE DATA ASSOCIATED WITH THE LOCATION OF THE USER.
• GOOGLE STREET VIEW API, TO PROVIDE THE USER WITH PHOTOS OF THE PLACES AND STREET LIKE VIEW FROM THE AIRPLANE.
• GEOGRAPHIC SATELLITE API, TO PROVIDE THE USER WITH THE GEOGRAPHIC DETAILS OF THE PLACE HE IS FLYING OVER.
• MACHINE LEARNING ALGORITHMS, TO LEARN THE NATURAL CHANGE OF THE EARTH SURFACE THROUGH THE TIME AND PREDICT WHAT WOULD HAPPEN IN THE FUTURE.
• MACHINE LEARNING ALGORITHMS, TO RECOMMEND THE USER PLACES WOULD BE OF HIS INTEREST.

WHAT IS THE DATA?

Geographic images from satellites presented in a hologram technology, and cultural locations of interest along their flight path, and the captured changes that have been in it, in the past, through the present, to the future.

HOW IT WILL BE DONE?

Front end

Responsive web application, to be accessed from any platform available out there in the market, it will use the precise position of the user, the location will be pushed every 5 seconds to the server to decrease the overhead on the device increasing its performance with the ability to manually push the location to the server, the application will be sectioned into geographic map and changes, history and cultural locations of interest, and natural locations and activities, there will be hologram display associated with the cultural locations of interest which the user will be able to see the location in a pop floating hologram scene.

Back end

Mainly we use Python language to handle the server side calculations, using APIs from NASA and Google to get the maps and locations around the position of the user, returning the information of the place with photos of it, not only in the present time but along the past all that happened in the history and was noticed by the satellites from changes, through the present and what we live in, going to the predicted future of the places using clustering and regression machine learning algorithms, which will be trained with the old observations to predict the future.

Machine Learning tasks

PREDICTING THE FUTURE OF THE PLACE:

Predicting the future of the place considering the green and environmental areas, the places captured from NASA provide a high quality images from satellites, by doing some ML algorithms between the change of images over specific interval in time using simple regression model to predict the future that will happen to that place.

RECOMMEND THE MOST RELATED INTEREST PLACES FOR THE VOYAGER:

Every day we have voyagers travel and visit other places, we make a recommender system using matrix factorization to make a personalized recommender model for every user for the application that provide some of recommended places that he can visit also.

We make a quality metric data visualization for two models done by recommendation one based on popularity of the place and other based on the personalization of the voyager which make a better recommendation specified for every voyager.

Visualized data for quality metric for recommender models

HOLOGRAM TASK

Hologram is an entertainment task to visualize the place in a 3D space in front of the voyager to know how it would be change over time, this task will be more developed in the future to make a real hologram presenter for the voyagers to see what they're flying on.

TECHNOLOGIES AND PROGRAMMING USED:

• "Graphlab create" and "SKlearn" python packages for Machine learning models.
• " Flask python framework " for deploying a back-end for the website and handling with ML models and handling the database creation and queries.
• JavaScript, HTML 5.0 and CSS 3.0 deploying a front-end for the website by uploading images and present the text and the place that received its coordinates from the API.

Resources:

Google's map API.

NASA Observatory Images.