dirzyadam/adajob

this for capstone project product-based

AdaJob

AdaJob is a mobile application to recommend Airdrop Cryptography jobs. In this application we will provide lists of available Airdrop jobs and give some Airdrop jobs recommendations to each user. To give recommendation to each user, we use Machine Learning to create the Collaborative Filtering Recommendation System. The tools used in this project are TensorFlow, TensorFlow Recommenders, and Google Colaboratory. This repository contains the main parts for the Machine Learning model as follows:

• Dataset Recommendation System : contains 2 csv files to train the model.
• Recommendation Model : contains Recommender_System.ipynb to train and export the model, Load_savedmodel.ipynb to test the exported model and get the recommendation prediction, and modelsavedwithkerasmeta.zip as the exported model file.

To duplicate our project, download the datasets first and run the Recommender_System.ipynb. The output will be modelsavedwithkerasmeta.zip. We refer to TensorFlow Recommenders' documentation to build our recommendation model.

Note:

For other machine learning models in this repository, we can add them in further developing AdaJob. For example, if later we want to develop user identity verification, whether it is to extract personal data from users based on OCR or to recognize user faces, it must first be ensured that the photo uploaded by the user is an id card or not. In addition, our model can focus more on the id card part only if it turns out that the uploaded photo contains objects that are not needed by the model.