TON 618

https://pastebin.com/NfrFtnbv

https://kiuge-my.sharepoint.com/:p:/g/personal/papava_elene_kiu_edu_ge/ETCoA6rl_TBCpzfeaByhhHsBwW6otXhdee3GuR-GA31SPw

Searching for specific topics in the NASA database is hard to find necessary information easily. The searching process needs time and human resources because the information in PDF files is unsearchable. The solution is our application (NASearch) which works with AI and improves the accessibility and discoverability of the records in the public NTRS. Our target audience is Scientists, researchers, students, and everyone who loves astrophysics, physics, and science. On the application, you can find relevant information for your project or research.

How does the Application work?

First of all, our application uses three main technics: Optical Character Recognition, Natural Language Processing, and Naive Bayes. We use OCR to convert handwritten or scanned PDF files into machine-encoded text. Secondly, we use NLP to analyze and statistic large amounts of natural language data. And lastly, we use Naive Bayes to allocate and classify data in the database. We work in Python using Tesseract and NLTK.

Also, the application uses an up-down voting system which recommends featured topics to improve customer experience.

We are five in a group: three of us are computer science students, one math student, and one management student. In our team we have diversity.

We use some PDF files from NASA's open source which was mentioned in challenge background.

While working with databases, searching was taking time and that's why we decided to come up with solution . We are the winners of local NASA KIU Hackathon from Georgia.

https://pastebin.com/NfrFtnbv

PDF was given by NASA.

#AI, #NASEARCH, #ML, #MachineLearning, #Python