/BerUang

Primary LanguageKotlin

BerUang

Manage Your Money With Us!

"BerUang" is a financial management application developed by our team to tackle the prevalent issue of online lending and poor financial decision-making. This app addresses the challenges by offering a monthly expense tracking feature that analyzes individual budgets, providing spending recommendations. Users receive notifications if they exceed predefined financial category budgets. Moreover, the app includes an article section that leverages machine learning to offer personalized financial guidance based on user behavior, enhancing financial planning capabilities. With these features, BerUang aims to empower users with better financial management practices and insightful guidance for improved monetary decision-making.

BerUang TEAM | CH2-PS111

Name Role Email
Ni Luh Ade Meina Rossalina Machine Learning m014bsx1813@bangkit.academy
Muhammad Akhsan Nurramdhan Machine Learning m267bsy1861@bangkit.academy
Mutiq Anisa Tanjung Machine Learning m296bsx0487@bangkit.academy
Lintang Arsa Naura Cloud Computing c227bsx3811@bangkit.academy
Ferrari Romano Cloud Computing c227bsy4033@bangkit.academy
Putri Dian Shafira Mobile Development a296bsx2745@bangkit.academy
Abdurrahman Al Aufa Mobile Development a283bsy2381@bangkit.academy

BerUang TEAM | Branch

Role Branch
Machine Learning ML
Cloud Computing api-artikel , backend-API
Mobile Development BerUang-App

Machine Learning

Overview

we create 2 machine learning models: (/n) -article recommendation system (/n) - article category classification

Dataset and Data Preparation The dataset we use is taken through scrapping several news potals including detik.com and bisnis.com. This training dataset is intended for the article category classification system. for this recommendation system is only a system of finding available articles with similar discussions through the cosine_similarity value.

This is the result of the dataset that we have processed: Dataset.

Example of Dataset with Each Labelsy
Model Summary

Model Result

Model Accuracy

Testing

Testing for article recommendation system
Testing for article category classification

Cloud Computing

This is backend-API for Beruang

Requirements

  • Text Editor (Visual Studio Code)
  • Git
  • Google Cloud Account
  • Google Cloud SDK

1. Setup Google Cloud Platfrom

Enable the following API :

  • App Engine API
  • Cloud Firestore API

2. Install Google Cloud SDK

3. Setup Firebase

  • Login to Firebase, go to console and connect it to Google Cloud Project.
  • Enable Firebase Authentication and Firestore Database.
  • Make a service account and download the file. (Make sure to integrate the serviceAccount.json with the code later)

4. Clone Project and set Google Cloud Account

5. Deploy the project on App Engine

  • Deploy the project to App Engine using 'gcloud app deploy'

6. API Documentation

You can check API Documentation in the following link. Postman API Beruang

This is Flask API Article for Beruang

This work is part of using article scraping techniques on news websites

This API has endpoints to make it work for the mobile development, it's integrated with Cloud Storage, The API is deployed on the Google Cloud Run.

Need

  • Python
  • Docker
  • Google Cloud Platform Account
  • Google Cloud Platform - Cloud Build API

Run locally

$ git clone -b api-article https://github.com/firareu/BerUang.git
$ pip install -r requirements. txt
$ python main.py

How to deploy to cloud-run using cloud SDK

$ gcloud init
$ gcloud services enable run.googleapis.com
$ gcloud builds submit --tag gcr.io/[project-id-kalian]/beruang-api-article
$ gcloud run deploy --image gcr.io/[project-id-kalian]/bear-fire-article --platform managed --region asia-southeast2 --allow-unauthenticated bear-fire-article

How to deploy to cloud-run using Google Cloud Platform

  1. Make sure you have an active Google Cloud Platform (GCP) account. If not, register and create a new project at https://console.cloud.google.com.

  2. Make sure you have Google Cloud SDK installed (https://cloud.google.com/sdk) and initialize it by running the following command in the terminal or command prompt: gcloud init

  3. Create a repository on a code management service such as GitHub or GitLab, and make sure it contains all the files required for your Flask application, including the Dockerfile, requirements.txt, and your Flask application code.

  4. Open a terminal or command prompt, then navigate to the directory where you want to clone the Flask repository.

  5. Clone the Flask repository by running the following command: git clone -b api-article https://github.com/firareu/BerUang.git

  6. Once the cloning process is complete, direct the terminal or command prompt to the Flask directory you just cloned.

  7. Build a local Docker container by running the following command: docker build -t gcr.io/[PROJECT_ID]/bear-api-articles . Replace [PROJECT_ID] with the Google Cloud Platform project ID that you specified previously.

  8. After the build process is complete, verify that the local Docker container is running by running the following command: docker run -p 8080:8080 gcr.io/[PROJECT_ID]/bear-api-article Make sure there are no errors and the Flask application is running fine on localhost.

  9. If the previous step was successful, stop and delete the running Docker container by pressing Ctrl+C in the terminal or command prompt.

  10. To publish the Docker container to Google Cloud Container Registry, run the following command: docker push gcr.io/[PROJECT_ID]/bear-fire-articles The container will be uploaded to the Container Registry in the appropriate Google Cloud Platform project.

  11. Next, create a Cloud Run service by running the following command on asia-southeast2(jakarta): gcloud run deploy --image gcr.io/[PROJECT_ID]/fire-bear-articles --platform managed --region asia-southeast2 --allow-unauthenticated fire-bears-articles

  12. GCP will ask you to select a region to deploy the Cloud Run service. Select the region that suits your needs.

  13. After the deployment process is complete, GCP will provide a URL that can be used to access the deployed Flask application. Copy the URL from the output and try accessing it in a web browser or by using API testing software such as Postman.

How to deploy to cloud-run using Google Cloud Platform

You can check API Documentation in the following link. Postman API Bear Article

Mobile Development

Overview

note: don't use darkmode

We build android applications for end users so that everyone can use our applications. The following is the application flow based on the Mockup.

Application Flowchart

And this is the libraries we used to build the android application:

Our application is not perfect where the application can only do the following things:

  • Login and register
  • Display articles
  • Provide article recommendations
  • And all the MVP UI of the application has been created

Displays using the following user:

Email: lynnarsa@example.com Pass: password123

  • Displays income data
  • Displays outcome data
  • Displays historical outcome data

Here is our application's Screenshot