High-Level Project Summary
The goal of our project is to raise awareness and interest in space. We decided to do this by creating a machine learning-based intelligent text classifier that performs automatic word classification of space-related sentences. The text entered by the user can be accepted by the computer. This is important because we are promoting knowledge about astronautics from official reports published by NASA researchers.
Link to Final Project
Link to Project "Demo"
Detailed Project Description
The goal of our project is to raise awareness and interest in space. We decided to do this by creating a machine learning-based intelligent text classifier that performs automatic word classification of space-related sentences. The computer can accept the text entered by the user. During the analysis of the words of the entered sentences, he will have to face various difficulties, such as the context of the words in relation to the sentence. He will need to make a quick decision in the face of many possible proposals. In the end, the user will receive information about the recognized words of the sentence. Moreover, in the future, we want to connect our intelligent text classifier with NASA text and graphic information databases. We can partner with various educational organizations to promote the idea that space is closer than you think and that working with official documents is fast and convenient.
Space Agency Data
Since our project is an intelligent text classifier created as a Python program, we used the data from the existing IMDB benchmark, which contains, among other things, space-related information. Our main goal is to make our project as realistic as possible - we try to convey information to our users in a clear and concise manner. We are interested in giving our users reliable and confirmed information and giving them the opportunity to improve their knowledge about space. Sci-fi films, which are now too popular, have given rise to many myths about astronautics and the activities of astronauts. Our goal is to classify the words of sentences considering the context conveniently and quickly When we wrote the intelligent text classifier and its tasks, we used the IMDB benchmark data, which contains, among other things, space-related information. The intelligent text classifier is based on a transformer that recognizes text based on context. NASA text data about various space technologies that need to be visualized inspired our project. We were so inspired by this idea that we chose the phrase for the slogan of our project “Recognize everything in our worlds”, the goal of which is to help masters classify the words of sentences for the purpose of subsequent visualization. We hope that our intelligent text classifier will help researchers fulfill their dreams and feel like real experts in the field of space technology.
Hackathon Journey
Our path to the hackathon was fast and very eventful. In just 48 hours, we came up with our own project almost from scratch and implemented it on our own. In the course of our work, we had to turn to NASA research data, study and systematize them, using previously acquired knowledge and obtaining new ones in the process. As a result of hard work, we have acquired new skills and improved old ones. We have learned to find optimal and non-standard solutions. We have learned to quickly adapt to new working conditions and use new, unfamiliar tools. Working with data from the NASA Technical Reports Server and other agencies has proven to be especially helpful, as we have improved our ability to intelligently classify data based on machine learning. In addition to purely scientific and technical skills, our team also went through the next stage of team building and solidarity. We were so imbued with the idea of searching for information about space technologies that are described by textual data that we chose a phrase for the slogan of our project “Recognize everything in our worlds”, the purpose of which is to help masters classify the words of sentences for the purpose of subsequent visualization.
References
Tags
#intellectual classifier #automatic classifier #text classifier #master #NASA #visualization information #space technology